 View on TensorFlow.org
|
 Run in Google Colab
|
 View on GitHub
|
 Download notebook
|
This tutorial fine-tunes a Mask R-CNN with Mobilenet V2 as backbone model from the TensorFlow Model Garden package (tensorflow-models).
Model Garden contains a collection of state-of-the-art models, implemented with TensorFlow's high-level APIs. The implementations demonstrate the best practices for modeling, letting users to take full advantage of TensorFlow for their research and product development.
This tutorial demonstrates how to:
- Use models from the TensorFlow Models package.
- Train/Fine-tune a pre-built Mask R-CNN with mobilenet as backbone for Object Detection and Instance Segmentation
- Export the trained/tuned Mask R-CNN model
Install Necessary Dependencies
pip install -U -q "tf-models-official"pip install -U -q remotezip tqdm opencv-python einops
Import required libraries
import os
import io
import json
import tqdm
import shutil
import pprint
import pathlib
import tempfile
import requests
import collections
import matplotlib
import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt
from PIL import Image
from six import BytesIO
from etils import epath
from IPython import display
from urllib.request import urlopen
import orbit
import tensorflow as tf
import tensorflow_models as tfm
import tensorflow_datasets as tfds
from official.core import exp_factory
from official.core import config_definitions as cfg
from official.vision.data import tfrecord_lib
from official.vision.serving import export_saved_model_lib
from official.vision.dataloaders.tf_example_decoder import TfExampleDecoder
from official.vision.utils.object_detection import visualization_utils
from official.vision.ops.preprocess_ops import normalize_image, resize_and_crop_image
from official.vision.data.create_coco_tf_record import coco_annotations_to_lists
pp = pprint.PrettyPrinter(indent=4) # Set Pretty Print Indentation
print(tf.__version__) # Check the version of tensorflow used
%matplotlib inline
/tmpfs/src/tf_docs_env/lib/python3.11/site-packages/tensorflow_hub/__init__.py:61: UserWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html. The pkg_resources package is slated for removal as early as 2025-11-30. Refrain from using this package or pin to Setuptools<81. from pkg_resources import parse_version 2.20.0
Download subset of lvis dataset
LVIS: A dataset for large vocabulary instance segmentation.
# @title Download annotation fileswget https://dl.fbaipublicfiles.com/LVIS/lvis_v1_train.json.zipunzip -q lvis_v1_train.json.ziprm lvis_v1_train.json.zipwget https://dl.fbaipublicfiles.com/LVIS/lvis_v1_val.json.zipunzip -q lvis_v1_val.json.ziprm lvis_v1_val.json.zipwget https://dl.fbaipublicfiles.com/LVIS/lvis_v1_image_info_test_dev.json.zipunzip -q lvis_v1_image_info_test_dev.json.ziprm lvis_v1_image_info_test_dev.json.zip
--2026-10-31 11:48:04-- https://dl.fbaipublicfiles.com/LVIS/lvis_v1_train.json.zip
Resolving dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)... 3.171.22.33, 3.171.22.118, 3.171.22.13, ...
Connecting to dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)|3.171.22.33|:443... connected.
HTTP request sent, awaiting response...
200 OK
Length: 350264821 (334M) [application/zip]
Saving to: ‘lvis_v1_train.json.zip’
lvis_v1_train.json. 0%[ ] 0 --.-KB/s
lvis_v1_train.json. 10%[=> ] 35.41M 177MB/s
lvis_v1_train.json. 25%[====> ] 83.75M 209MB/s
lvis_v1_train.json. 39%[======> ] 132.01M 220MB/s
lvis_v1_train.json. 54%[=========> ] 182.74M 228MB/s
lvis_v1_train.json. 70%[=============> ] 233.99M 234MB/s
lvis_v1_train.json. 85%[================> ] 286.03M 238MB/s
lvis_v1_train.json. 100%[===================>] 334.04M 241MB/s in 1.4s
2026-10-31 11:48:05 (241 MB/s) - ‘lvis_v1_train.json.zip’ saved [350264821/350264821]
--2026-10-31 11:48:14-- https://dl.fbaipublicfiles.com/LVIS/lvis_v1_val.json.zip
Resolving dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)... 3.171.22.33, 3.171.22.118, 3.171.22.13, ...
Connecting to dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)|3.171.22.33|:443... connected.
HTTP request sent, awaiting response...
200 OK
Length: 64026968 (61M) [application/zip]
Saving to: ‘lvis_v1_val.json.zip’
lvis_v1_val.json.zi 0%[ ] 0 --.-KB/s
lvis_v1_val.json.zi 55%[==========> ] 33.98M 170MB/s
lvis_v1_val.json.zi 100%[===================>] 61.06M 200MB/s in 0.3s
2026-10-31 11:48:15 (200 MB/s) - ‘lvis_v1_val.json.zip’ saved [64026968/64026968]
--2026-10-31 11:48:17-- https://dl.fbaipublicfiles.com/LVIS/lvis_v1_image_info_test_dev.json.zip
Resolving dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)... 3.171.22.33, 3.171.22.118, 3.171.22.13, ...
Connecting to dl.fbaipublicfiles.com (dl.fbaipublicfiles.com)|3.171.22.33|:443... connected.
HTTP request sent, awaiting response...
200 OK
Length: 384629 (376K) [application/zip]
Saving to: ‘lvis_v1_image_info_test_dev.json.zip’
lvis_v1_i 0%[ ] 0 --.-KB/s
lvis_v1_image_info_ 100%[===================>] 375.61K 2.02MB/s in 0.2s
2026-10-31 11:48:17 (2.02 MB/s) - ‘lvis_v1_image_info_test_dev.json.zip’ saved [384629/384629]
# @title Lvis annotation parsing
# Annotations with invalid bounding boxes. Will not be used.
_INVALID_ANNOTATIONS = [
# Train split.
662101,
81217,
462924,
227817,
29381,
601484,
412185,
504667,
572573,
91937,
239022,
181534,
101685,
# Validation split.
36668,
57541,
33126,
10932,
]
def get_category_map(annotation_path, num_classes):
with epath.Path(annotation_path).open() as f:
data = json.load(f)
category_map = {id+1: {'id': cat_dict['id'],
'name': cat_dict['name']}
for id, cat_dict in enumerate(data['categories'][:num_classes])}
return category_map
class LvisAnnotation:
"""LVIS annotation helper class.
The format of the annations is explained on
https://www.lvisdataset.org/dataset.
"""
def __init__(self, annotation_path):
with epath.Path(annotation_path).open() as f:
data = json.load(f)
self._data = data
img_id2annotations = collections.defaultdict(list)
for a in self._data.get('annotations', []):
if a['category_id'] in category_ids:
img_id2annotations[a['image_id']].append(a)
self._img_id2annotations = {
k: list(sorted(v, key=lambda a: a['id']))
for k, v in img_id2annotations.items()
}
@property
def categories(self):
"""Return the category dicts, as sorted in the file."""
return self._data['categories']
@property
def images(self):
"""Return the image dicts, as sorted in the file."""
sub_images = []
for image_info in self._data['images']:
if image_info['id'] in self._img_id2annotations:
sub_images.append(image_info)
return sub_images
def get_annotations(self, img_id):
"""Return all annotations associated with the image id string."""
# Some images don't have any annotations. Return empty list instead.
return self._img_id2annotations.get(img_id, [])
def _generate_tf_records(prefix, images_zip, annotation_file, num_shards=5):
"""Generate TFRecords."""
lvis_annotation = LvisAnnotation(annotation_file)
def _process_example(prefix, image_info, id_to_name_map):
# Search image dirs.
filename = pathlib.Path(image_info['coco_url']).name
image = tf.io.read_file(os.path.join(IMGS_DIR, filename))
instances = lvis_annotation.get_annotations(img_id=image_info['id'])
instances = [x for x in instances if x['id'] not in _INVALID_ANNOTATIONS]
# print([x['category_id'] for x in instances])
is_crowd = {'iscrowd': 0}
instances = [dict(x, **is_crowd) for x in instances]
neg_category_ids = image_info.get('neg_category_ids', [])
not_exhaustive_category_ids = image_info.get(
'not_exhaustive_category_ids', []
)
data, _ = coco_annotations_to_lists(instances,
id_to_name_map,
image_info['height'],
image_info['width'],
include_masks=True)
# data['category_id'] = [id-1 for id in data['category_id']]
keys_to_features = {
'image/encoded':
tfrecord_lib.convert_to_feature(image.numpy()),
'image/filename':
tfrecord_lib.convert_to_feature(filename.encode('utf8')),
'image/format':
tfrecord_lib.convert_to_feature('jpg'.encode('utf8')),
'image/height':
tfrecord_lib.convert_to_feature(image_info['height']),
'image/width':
tfrecord_lib.convert_to_feature(image_info['width']),
'image/source_id':
tfrecord_lib.convert_to_feature(str(image_info['id']).encode('utf8')),
'image/object/bbox/xmin':
tfrecord_lib.convert_to_feature(data['xmin']),
'image/object/bbox/xmax':
tfrecord_lib.convert_to_feature(data['xmax']),
'image/object/bbox/ymin':
tfrecord_lib.convert_to_feature(data['ymin']),
'image/object/bbox/ymax':
tfrecord_lib.convert_to_feature(data['ymax']),
'image/object/class/text':
tfrecord_lib.convert_to_feature(data['category_names']),
'image/object/class/label':
tfrecord_lib.convert_to_feature(data['category_id']),
'image/object/is_crowd':
tfrecord_lib.convert_to_feature(data['is_crowd']),
'image/object/area':
tfrecord_lib.convert_to_feature(data['area'], 'float_list'),
'image/object/mask':
tfrecord_lib.convert_to_feature(data['encoded_mask_png'])
}
# print(keys_to_features['image/object/class/label'])
example = tf.train.Example(
features=tf.train.Features(feature=keys_to_features))
return example
# file_names = [f"{prefix}/{pathlib.Path(image_info['coco_url']).name}"
# for image_info in lvis_annotation.images]
# _extract_images(images_zip, file_names)
writers = [
tf.io.TFRecordWriter(
tf_records_dir + prefix +'-%05d-of-%05d.tfrecord' % (i, num_shards))
for i in range(num_shards)
]
id_to_name_map = {cat_dict['id']: cat_dict['name']
for cat_dict in lvis_annotation.categories[:NUM_CLASSES]}
# print(id_to_name_map)
for idx, image_info in enumerate(tqdm.tqdm(lvis_annotation.images)):
img_data = requests.get(image_info['coco_url'], stream=True).content
img_name = image_info['coco_url'].split('/')[-1]
with open(os.path.join(IMGS_DIR, img_name), 'wb') as handler:
handler.write(img_data)
tf_example = _process_example(prefix, image_info, id_to_name_map)
writers[idx % num_shards].write(tf_example.SerializeToString())
del lvis_annotation
_URLS = {
'train_images': 'http://images.cocodataset.org/zips/train2017.zip',
'validation_images': 'http://images.cocodataset.org/zips/val2017.zip',
'test_images': 'http://images.cocodataset.org/zips/test2017.zip',
}
train_prefix = 'train'
valid_prefix = 'val'
train_annotation_path = './lvis_v1_train.json'
valid_annotation_path = './lvis_v1_val.json'
IMGS_DIR = './lvis_sub_dataset/'
tf_records_dir = './lvis_tfrecords/'
if not os.path.exists(IMGS_DIR):
os.mkdir(IMGS_DIR)
if not os.path.exists(tf_records_dir):
os.mkdir(tf_records_dir)
NUM_CLASSES = 3
category_index = get_category_map(valid_annotation_path, NUM_CLASSES)
category_ids = list(category_index.keys())
# Below helper function are taken from github tensorflow dataset lvis
# https://github.com/tensorflow/datasets/blob/master/tensorflow_datasets/datasets/lvis/lvis_dataset_builder.py
_generate_tf_records(train_prefix,
_URLS['train_images'],
train_annotation_path)
0%| | 0/2338 [00:00<?, ?it/s] WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1793447340.873058 41008 gpu_device.cc:2020] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 13638 MB memory: -> device: 0, name: Tesla T4, pci bus id: 0000:00:05.0, compute capability: 7.5 I0000 00:00:1793447340.875508 41008 gpu_device.cc:2020] Created device /job:localhost/replica:0/task:0/device:GPU:1 with 13756 MB memory: -> device: 1, name: Tesla T4, pci bus id: 0000:00:06.0, compute capability: 7.5 I0000 00:00:1793447340.877793 41008 gpu_device.cc:2020] Created device /job:localhost/replica:0/task:0/device:GPU:2 with 13756 MB memory: -> device: 2, name: Tesla T4, pci bus id: 0000:00:07.0, compute capability: 7.5 I0000 00:00:1793447340.879958 41008 gpu_device.cc:2020] Created device /job:localhost/replica:0/task:0/device:GPU:3 with 13756 MB memory: -> device: 3, name: Tesla T4, pci bus id: 0000:00:08.0, compute capability: 7.5 0%| | 1/2338 [00:02<1:37:07, 2.49s/it] 0%| | 2/2338 [00:02<45:04, 1.16s/it] 0%| | 3/2338 [00:02<28:28, 1.37it/s] 0%| | 4/2338 [00:03<20:39, 1.88it/s] 0%| | 5/2338 [00:03<16:30, 2.35it/s] 0%| | 6/2338 [00:03<14:25, 2.69it/s] 0%| | 7/2338 [00:03<13:10, 2.95it/s] 0%| | 8/2338 [00:04<11:45, 3.30it/s] 0%| | 9/2338 [00:04<10:28, 3.70it/s] 0%| | 10/2338 [00:04<08:53, 4.37it/s] 0%| | 11/2338 [00:04<08:16, 4.68it/s] 1%| | 12/2338 [00:04<08:21, 4.63it/s] 1%| | 13/2338 [00:05<08:36, 4.50it/s] 1%| | 14/2338 [00:05<09:10, 4.22it/s] 1%| | 15/2338 [00:05<08:47, 4.40it/s] 1%| | 16/2338 [00:05<09:25, 4.11it/s] 1%| | 17/2338 [00:06<09:04, 4.26it/s] 1%| | 18/2338 [00:06<09:09, 4.22it/s] 1%| | 19/2338 [00:06<09:27, 4.08it/s] 1%| | 20/2338 [00:06<09:15, 4.17it/s] 1%| | 21/2338 [00:07<09:06, 4.24it/s] 1%| | 22/2338 [00:07<09:01, 4.27it/s] 1%| | 23/2338 [00:07<08:33, 4.51it/s] 1%| | 24/2338 [00:07<08:31, 4.52it/s] 1%| | 25/2338 [00:07<09:02, 4.26it/s] 1%| | 26/2338 [00:08<09:17, 4.15it/s] 1%| | 27/2338 [00:08<09:24, 4.10it/s] 1%| | 28/2338 [00:08<09:10, 4.19it/s] 1%| | 29/2338 [00:08<09:24, 4.09it/s] 1%|▏ | 30/2338 [00:09<09:13, 4.17it/s] 1%|▏ | 31/2338 [00:09<08:14, 4.67it/s] 1%|▏ | 32/2338 [00:09<08:19, 4.62it/s] 1%|▏ | 33/2338 [00:09<08:23, 4.58it/s] 1%|▏ | 34/2338 [00:10<08:19, 4.61it/s] 1%|▏ | 35/2338 [00:10<08:30, 4.51it/s] 2%|▏ | 36/2338 [00:10<08:58, 4.27it/s] 2%|▏ | 37/2338 [00:10<08:01, 4.78it/s] 2%|▏ | 38/2338 [00:10<08:14, 4.65it/s] 2%|▏ | 39/2338 [00:11<08:09, 4.69it/s] 2%|▏ | 40/2338 [00:11<08:33, 4.47it/s] 2%|▏ | 41/2338 [00:11<08:43, 4.39it/s] 2%|▏ | 42/2338 [00:11<08:06, 4.72it/s] 2%|▏ | 43/2338 [00:11<07:48, 4.90it/s] 2%|▏ | 44/2338 [00:12<08:22, 4.56it/s] 2%|▏ | 45/2338 [00:12<08:32, 4.48it/s] 2%|▏ | 46/2338 [00:12<08:18, 4.60it/s] 2%|▏ | 47/2338 [00:12<08:10, 4.67it/s] 2%|▏ | 48/2338 [00:13<08:25, 4.53it/s] 2%|▏ | 49/2338 [00:13<08:06, 4.71it/s] 2%|▏ | 50/2338 [00:13<08:19, 4.58it/s] 2%|▏ | 51/2338 [00:13<08:22, 4.55it/s] 2%|▏ | 52/2338 [00:13<08:37, 4.42it/s] 2%|▏ | 53/2338 [00:14<08:48, 4.32it/s] 2%|▏ | 54/2338 [00:14<09:20, 4.08it/s] 2%|▏ | 55/2338 [00:14<09:03, 4.20it/s] 2%|▏ | 56/2338 [00:14<08:34, 4.44it/s] 2%|▏ | 57/2338 [00:15<08:49, 4.31it/s] 2%|▏ | 58/2338 [00:15<08:54, 4.26it/s] 3%|▎ | 59/2338 [00:15<08:50, 4.29it/s] 3%|▎ | 60/2338 [00:15<08:50, 4.29it/s] 3%|▎ | 61/2338 [00:16<08:55, 4.25it/s] 3%|▎ | 62/2338 [00:16<08:56, 4.24it/s] 3%|▎ | 63/2338 [00:16<09:16, 4.09it/s] 3%|▎ | 64/2338 [00:16<09:06, 4.16it/s] 3%|▎ | 65/2338 [00:17<09:02, 4.19it/s] 3%|▎ | 66/2338 [00:17<09:08, 4.14it/s] 3%|▎ | 67/2338 [00:17<08:56, 4.24it/s] 3%|▎ | 68/2338 [00:17<08:44, 4.33it/s] 3%|▎ | 69/2338 [00:17<08:46, 4.31it/s] 3%|▎ | 70/2338 [00:18<08:33, 4.42it/s] 3%|▎ | 71/2338 [00:18<08:35, 4.39it/s] 3%|▎ | 72/2338 [00:18<09:10, 4.12it/s] 3%|▎ | 73/2338 [00:18<09:04, 4.16it/s] 3%|▎ | 74/2338 [00:19<09:33, 3.95it/s] 3%|▎ | 75/2338 [00:19<09:19, 4.04it/s] 3%|▎ | 76/2338 [00:19<09:23, 4.02it/s] 3%|▎ | 77/2338 [00:19<09:10, 4.11it/s] 3%|▎ | 78/2338 [00:20<09:12, 4.09it/s] 3%|▎ | 79/2338 [00:20<09:25, 4.00it/s] 3%|▎ | 80/2338 [00:20<09:00, 4.18it/s] 3%|▎ | 81/2338 [00:20<08:31, 4.41it/s] 4%|▎ | 82/2338 [00:21<08:10, 4.60it/s] 4%|▎ | 83/2338 [00:21<08:05, 4.65it/s] 4%|▎ | 84/2338 [00:21<07:32, 4.98it/s] 4%|▎ | 85/2338 [00:21<08:35, 4.37it/s] 4%|▎ | 86/2338 [00:21<08:44, 4.29it/s] 4%|▎ | 87/2338 [00:22<08:43, 4.30it/s] 4%|▍ | 88/2338 [00:22<08:47, 4.26it/s] 4%|▍ | 89/2338 [00:22<08:32, 4.39it/s] 4%|▍ | 90/2338 [00:22<09:00, 4.16it/s] 4%|▍ | 91/2338 [00:23<09:10, 4.09it/s] 4%|▍ | 92/2338 [00:23<08:51, 4.23it/s] 4%|▍ | 93/2338 [00:23<08:43, 4.29it/s] 4%|▍ | 94/2338 [00:23<08:45, 4.27it/s] 4%|▍ | 95/2338 [00:24<08:32, 4.38it/s] 4%|▍ | 96/2338 [00:24<07:47, 4.79it/s] 4%|▍ | 97/2338 [00:24<08:18, 4.50it/s] 4%|▍ | 98/2338 [00:24<08:24, 4.44it/s] 4%|▍ | 99/2338 [00:24<08:57, 4.16it/s] 4%|▍ | 100/2338 [00:25<08:58, 4.15it/s] 4%|▍ | 101/2338 [00:25<08:42, 4.28it/s] 4%|▍ | 102/2338 [00:25<08:29, 4.39it/s] 4%|▍ | 103/2338 [00:25<08:31, 4.37it/s] 4%|▍ | 104/2338 [00:26<08:51, 4.20it/s] 4%|▍ | 105/2338 [00:26<08:42, 4.27it/s] 5%|▍ | 106/2338 [00:26<08:50, 4.21it/s] 5%|▍ | 107/2338 [00:26<08:31, 4.36it/s] 5%|▍ | 108/2338 [00:27<08:33, 4.34it/s] 5%|▍ | 109/2338 [00:27<08:33, 4.34it/s] 5%|▍ | 110/2338 [00:27<09:03, 4.10it/s] 5%|▍ | 111/2338 [00:27<09:00, 4.12it/s] 5%|▍ | 112/2338 [00:28<09:19, 3.98it/s] 5%|▍ | 113/2338 [00:28<08:57, 4.14it/s] 5%|▍ | 114/2338 [00:28<07:59, 4.64it/s] 5%|▍ | 115/2338 [00:28<08:29, 4.36it/s] 5%|▍ | 116/2338 [00:28<08:49, 4.19it/s] 5%|▌ | 117/2338 [00:29<08:19, 4.45it/s] 5%|▌ | 118/2338 [00:29<08:19, 4.44it/s] 5%|▌ | 119/2338 [00:29<08:31, 4.34it/s] 5%|▌ | 120/2338 [00:29<08:42, 4.24it/s] 5%|▌ | 121/2338 [00:30<08:33, 4.32it/s] 5%|▌ | 122/2338 [00:30<08:34, 4.31it/s] 5%|▌ | 123/2338 [00:30<08:25, 4.38it/s] 5%|▌ | 124/2338 [00:30<08:25, 4.38it/s] 5%|▌ | 125/2338 [00:31<08:42, 4.24it/s] 5%|▌ | 126/2338 [00:31<08:39, 4.26it/s] 5%|▌ | 127/2338 [00:31<08:51, 4.16it/s] 5%|▌ | 128/2338 [00:31<08:41, 4.24it/s] 6%|▌ | 129/2338 [00:32<09:19, 3.95it/s] 6%|▌ | 130/2338 [00:32<09:12, 4.00it/s] 6%|▌ | 131/2338 [00:32<08:35, 4.28it/s] 6%|▌ | 132/2338 [00:32<08:38, 4.25it/s] 6%|▌ | 133/2338 [00:32<08:29, 4.33it/s] 6%|▌ | 134/2338 [00:33<08:19, 4.41it/s] 6%|▌ | 135/2338 [00:33<08:12, 4.47it/s] 6%|▌ | 136/2338 [00:33<08:14, 4.45it/s] 6%|▌ | 137/2338 [00:33<08:39, 4.24it/s] 6%|▌ | 138/2338 [00:34<09:00, 4.07it/s] 6%|▌ | 139/2338 [00:34<08:52, 4.13it/s] 6%|▌ | 140/2338 [00:34<09:03, 4.04it/s] 6%|▌ | 141/2338 [00:34<08:43, 4.19it/s] 6%|▌ | 142/2338 [00:35<08:54, 4.11it/s] 6%|▌ | 143/2338 [00:35<09:29, 3.85it/s] 6%|▌ | 144/2338 [00:35<09:32, 3.83it/s] 6%|▌ | 145/2338 [00:35<09:16, 3.94it/s] 6%|▌ | 146/2338 [00:36<08:22, 4.36it/s] 6%|▋ | 147/2338 [00:36<08:15, 4.42it/s] 6%|▋ | 148/2338 [00:36<08:47, 4.16it/s] 6%|▋ | 149/2338 [00:36<08:41, 4.20it/s] 6%|▋ | 150/2338 [00:37<09:26, 3.86it/s] 6%|▋ | 151/2338 [00:37<09:51, 3.70it/s] 7%|▋ | 152/2338 [00:37<09:09, 3.97it/s] 7%|▋ | 153/2338 [00:37<08:55, 4.08it/s] 7%|▋ | 154/2338 [00:38<08:54, 4.09it/s] 7%|▋ | 155/2338 [00:38<08:48, 4.13it/s] 7%|▋ | 156/2338 [00:38<08:52, 4.10it/s] 7%|▋ | 157/2338 [00:38<08:55, 4.08it/s] 7%|▋ | 158/2338 [00:39<08:41, 4.18it/s] 7%|▋ | 159/2338 [00:39<08:47, 4.13it/s] 7%|▋ | 160/2338 [00:39<08:18, 4.37it/s] 7%|▋ | 161/2338 [00:39<08:32, 4.25it/s] 7%|▋ | 162/2338 [00:39<08:13, 4.41it/s] 7%|▋ | 163/2338 [00:40<08:22, 4.33it/s] 7%|▋ | 164/2338 [00:40<08:25, 4.30it/s] 7%|▋ | 165/2338 [00:40<08:15, 4.38it/s] 7%|▋ | 166/2338 [00:40<08:16, 4.38it/s] 7%|▋ | 167/2338 [00:41<08:19, 4.35it/s] 7%|▋ | 168/2338 [00:41<08:38, 4.19it/s] 7%|▋ | 169/2338 [00:41<08:44, 4.14it/s] 7%|▋ | 170/2338 [00:41<09:03, 3.99it/s] 7%|▋ | 171/2338 [00:42<08:15, 4.38it/s] 7%|▋ | 172/2338 [00:42<08:35, 4.20it/s] 7%|▋ | 173/2338 [00:42<08:29, 4.25it/s] 7%|▋ | 174/2338 [00:42<08:47, 4.10it/s] 7%|▋ | 175/2338 [00:43<08:26, 4.27it/s] 8%|▊ | 176/2338 [00:43<08:26, 4.27it/s] 8%|▊ | 177/2338 [00:43<09:03, 3.98it/s] 8%|▊ | 178/2338 [00:43<09:08, 3.94it/s] 8%|▊ | 179/2338 [00:44<08:33, 4.21it/s] 8%|▊ | 180/2338 [00:44<07:35, 4.73it/s] 8%|▊ | 181/2338 [00:44<08:14, 4.36it/s] 8%|▊ | 182/2338 [00:44<08:24, 4.28it/s] 8%|▊ | 183/2338 [00:44<08:37, 4.16it/s] 8%|▊ | 184/2338 [00:45<08:43, 4.11it/s] 8%|▊ | 185/2338 [00:45<08:45, 4.10it/s] 8%|▊ | 186/2338 [00:45<08:42, 4.12it/s] 8%|▊ | 187/2338 [00:45<08:32, 4.19it/s] 8%|▊ | 188/2338 [00:46<08:30, 4.21it/s] 8%|▊ | 189/2338 [00:46<08:38, 4.14it/s] 8%|▊ | 190/2338 [00:46<08:26, 4.24it/s] 8%|▊ | 191/2338 [00:46<08:35, 4.16it/s] 8%|▊ | 192/2338 [00:47<08:47, 4.07it/s] 8%|▊ | 193/2338 [00:47<09:11, 3.89it/s] 8%|▊ | 194/2338 [00:47<08:44, 4.09it/s] 8%|▊ | 195/2338 [00:47<08:22, 4.26it/s] 8%|▊ | 196/2338 [00:48<07:53, 4.53it/s] 8%|▊ | 197/2338 [00:48<07:19, 4.87it/s] 8%|▊ | 198/2338 [00:48<08:02, 4.44it/s] 9%|▊ | 199/2338 [00:48<08:23, 4.25it/s] 9%|▊ | 200/2338 [00:48<07:47, 4.58it/s] 9%|▊ | 201/2338 [00:49<07:33, 4.71it/s] 9%|▊ | 202/2338 [00:49<07:49, 4.54it/s] 9%|▊ | 203/2338 [00:49<08:26, 4.21it/s] 9%|▊ | 204/2338 [00:49<08:19, 4.28it/s] 9%|▉ | 205/2338 [00:50<08:37, 4.12it/s] 9%|▉ | 206/2338 [00:50<08:40, 4.10it/s] 9%|▉ | 207/2338 [00:50<08:38, 4.11it/s] 9%|▉ | 208/2338 [00:50<08:10, 4.34it/s] 9%|▉ | 209/2338 [00:51<08:34, 4.14it/s] 9%|▉ | 210/2338 [00:51<08:01, 4.42it/s] 9%|▉ | 211/2338 [00:51<07:57, 4.46it/s] 9%|▉ | 212/2338 [00:51<08:03, 4.39it/s] 9%|▉ | 213/2338 [00:51<08:09, 4.34it/s] 9%|▉ | 214/2338 [00:52<07:57, 4.45it/s] 9%|▉ | 215/2338 [00:52<07:43, 4.58it/s] 9%|▉ | 216/2338 [00:52<08:01, 4.40it/s] 9%|▉ | 217/2338 [00:52<07:36, 4.64it/s] 9%|▉ | 218/2338 [00:53<07:52, 4.49it/s] 9%|▉ | 219/2338 [00:53<07:34, 4.66it/s] 9%|▉ | 220/2338 [00:53<07:39, 4.61it/s] 9%|▉ | 221/2338 [00:53<08:01, 4.40it/s] 9%|▉ | 222/2338 [00:53<08:06, 4.35it/s] 10%|▉ | 223/2338 [00:54<08:10, 4.31it/s] 10%|▉ | 224/2338 [00:54<08:22, 4.21it/s] 10%|▉ | 225/2338 [00:54<08:21, 4.21it/s] 10%|▉ | 226/2338 [00:54<08:12, 4.29it/s] 10%|▉ | 227/2338 [00:55<08:53, 3.96it/s] 10%|▉ | 228/2338 [00:55<08:58, 3.92it/s] 10%|▉ | 229/2338 [00:55<09:04, 3.87it/s] 10%|▉ | 230/2338 [00:55<09:14, 3.80it/s] 10%|▉ | 231/2338 [00:56<09:06, 3.86it/s] 10%|▉ | 232/2338 [00:56<08:41, 4.04it/s] 10%|▉ | 233/2338 [00:56<07:45, 4.52it/s] 10%|█ | 234/2338 [00:56<07:30, 4.67it/s] 10%|█ | 235/2338 [00:57<07:35, 4.61it/s] 10%|█ | 236/2338 [00:57<07:42, 4.55it/s] 10%|█ | 237/2338 [00:57<07:04, 4.95it/s] 10%|█ | 238/2338 [00:57<06:36, 5.30it/s] 10%|█ | 239/2338 [00:57<06:43, 5.20it/s] 10%|█ | 240/2338 [00:58<07:13, 4.84it/s] 10%|█ | 241/2338 [00:58<07:27, 4.69it/s] 10%|█ | 242/2338 [00:58<07:54, 4.42it/s] 10%|█ | 243/2338 [00:58<08:00, 4.36it/s] 10%|█ | 244/2338 [00:59<08:16, 4.21it/s] 10%|█ | 245/2338 [00:59<08:35, 4.06it/s] 11%|█ | 246/2338 [00:59<08:27, 4.12it/s] 11%|█ | 247/2338 [00:59<08:10, 4.26it/s] 11%|█ | 248/2338 [00:59<08:05, 4.31it/s] 11%|█ | 249/2338 [01:00<07:58, 4.36it/s] 11%|█ | 250/2338 [01:00<07:39, 4.55it/s] 11%|█ | 251/2338 [01:00<07:25, 4.69it/s] 11%|█ | 252/2338 [01:00<07:05, 4.91it/s] 11%|█ | 253/2338 [01:00<06:46, 5.13it/s] 11%|█ | 254/2338 [01:01<06:36, 5.26it/s] 11%|█ | 255/2338 [01:01<07:00, 4.95it/s] 11%|█ | 256/2338 [01:01<07:13, 4.81it/s] 11%|█ | 257/2338 [01:01<07:24, 4.68it/s] 11%|█ | 258/2338 [01:02<07:40, 4.52it/s] 11%|█ | 259/2338 [01:02<07:39, 4.52it/s] 11%|█ | 260/2338 [01:02<07:56, 4.36it/s] 11%|█ | 261/2338 [01:02<07:54, 4.37it/s] 11%|█ | 262/2338 [01:02<07:36, 4.54it/s] 11%|█ | 263/2338 [01:03<07:56, 4.36it/s] 11%|█▏ | 264/2338 [01:03<08:12, 4.21it/s] 11%|█▏ | 265/2338 [01:03<08:24, 4.11it/s] 11%|█▏ | 266/2338 [01:03<08:21, 4.13it/s] 11%|█▏ | 267/2338 [01:04<08:10, 4.22it/s] 11%|█▏ | 268/2338 [01:04<07:58, 4.33it/s] 12%|█▏ | 269/2338 [01:04<07:49, 4.41it/s] 12%|█▏ | 270/2338 [01:04<07:40, 4.49it/s] 12%|█▏ | 271/2338 [01:05<07:45, 4.44it/s] 12%|█▏ | 272/2338 [01:05<07:40, 4.48it/s] 12%|█▏ | 273/2338 [01:05<07:33, 4.55it/s] 12%|█▏ | 274/2338 [01:05<07:43, 4.46it/s] 12%|█▏ | 275/2338 [01:05<08:01, 4.28it/s] 12%|█▏ | 276/2338 [01:06<08:09, 4.21it/s] 12%|█▏ | 277/2338 [01:06<08:10, 4.20it/s] 12%|█▏ | 278/2338 [01:06<08:13, 4.17it/s] 12%|█▏ | 279/2338 [01:06<08:05, 4.24it/s] 12%|█▏ | 280/2338 [01:07<08:08, 4.21it/s] 12%|█▏ | 281/2338 [01:07<08:14, 4.16it/s] 12%|█▏ | 282/2338 [01:07<08:26, 4.06it/s] 12%|█▏ | 283/2338 [01:07<08:21, 4.10it/s] 12%|█▏ | 284/2338 [01:08<07:23, 4.63it/s] 12%|█▏ | 285/2338 [01:08<07:27, 4.59it/s] 12%|█▏ | 286/2338 [01:08<07:27, 4.59it/s] 12%|█▏ | 287/2338 [01:08<07:35, 4.51it/s] 12%|█▏ | 288/2338 [01:08<07:57, 4.29it/s] 12%|█▏ | 289/2338 [01:09<07:51, 4.34it/s] 12%|█▏ | 290/2338 [01:09<08:17, 4.11it/s] 12%|█▏ | 291/2338 [01:09<08:20, 4.09it/s] 12%|█▏ | 292/2338 [01:09<08:07, 4.20it/s] 13%|█▎ | 293/2338 [01:10<08:01, 4.25it/s] 13%|█▎ | 294/2338 [01:10<08:03, 4.22it/s] 13%|█▎ | 295/2338 [01:10<08:04, 4.22it/s] 13%|█▎ | 296/2338 [01:10<07:56, 4.29it/s] 13%|█▎ | 297/2338 [01:11<08:38, 3.93it/s] 13%|█▎ | 298/2338 [01:11<08:02, 4.23it/s] 13%|█▎ | 299/2338 [01:11<07:56, 4.28it/s] 13%|█▎ | 300/2338 [01:11<07:46, 4.37it/s] 13%|█▎ | 301/2338 [01:12<07:38, 4.45it/s] 13%|█▎ | 302/2338 [01:12<07:33, 4.49it/s] 13%|█▎ | 303/2338 [01:12<07:34, 4.47it/s] 13%|█▎ | 304/2338 [01:12<07:51, 4.31it/s] 13%|█▎ | 305/2338 [01:12<07:45, 4.37it/s] 13%|█▎ | 306/2338 [01:13<07:14, 4.68it/s] 13%|█▎ | 307/2338 [01:13<07:07, 4.76it/s] 13%|█▎ | 308/2338 [01:13<07:23, 4.58it/s] 13%|█▎ | 309/2338 [01:13<07:23, 4.58it/s] 13%|█▎ | 310/2338 [01:13<07:24, 4.56it/s] 13%|█▎ | 311/2338 [01:14<07:16, 4.65it/s] 13%|█▎ | 312/2338 [01:14<07:16, 4.64it/s] 13%|█▎ | 313/2338 [01:14<07:11, 4.69it/s] 13%|█▎ | 314/2338 [01:14<07:59, 4.22it/s] 13%|█▎ | 315/2338 [01:15<08:16, 4.07it/s] 14%|█▎ | 316/2338 [01:15<08:14, 4.09it/s] 14%|█▎ | 317/2338 [01:15<08:17, 4.07it/s] 14%|█▎ | 318/2338 [01:15<08:17, 4.06it/s] 14%|█▎ | 319/2338 [01:16<08:17, 4.06it/s] 14%|█▎ | 320/2338 [01:16<08:30, 3.95it/s] 14%|█▎ | 321/2338 [01:16<07:53, 4.26it/s] 14%|█▍ | 322/2338 [01:16<07:46, 4.32it/s] 14%|█▍ | 323/2338 [01:17<07:35, 4.42it/s] 14%|█▍ | 324/2338 [01:17<08:11, 4.10it/s] 14%|█▍ | 325/2338 [01:17<08:01, 4.18it/s] 14%|█▍ | 326/2338 [01:17<07:45, 4.32it/s] 14%|█▍ | 327/2338 [01:18<07:37, 4.40it/s] 14%|█▍ | 328/2338 [01:18<07:41, 4.35it/s] 14%|█▍ | 329/2338 [01:18<08:16, 4.05it/s] 14%|█▍ | 330/2338 [01:18<07:55, 4.22it/s] 14%|█▍ | 331/2338 [01:18<08:01, 4.17it/s] 14%|█▍ | 332/2338 [01:19<07:32, 4.44it/s] 14%|█▍ | 333/2338 [01:19<07:17, 4.59it/s] 14%|█▍ | 334/2338 [01:19<07:38, 4.37it/s] 14%|█▍ | 335/2338 [01:19<07:39, 4.36it/s] 14%|█▍ | 336/2338 [01:20<08:14, 4.05it/s] 14%|█▍ | 337/2338 [01:20<08:23, 3.98it/s] 14%|█▍ | 338/2338 [01:20<08:12, 4.06it/s] 14%|█▍ | 339/2338 [01:20<07:57, 4.19it/s] 15%|█▍ | 340/2338 [01:21<07:24, 4.50it/s] 15%|█▍ | 341/2338 [01:21<07:41, 4.32it/s] 15%|█▍ | 342/2338 [01:21<07:44, 4.30it/s] 15%|█▍ | 343/2338 [01:21<07:46, 4.28it/s] 15%|█▍ | 344/2338 [01:22<07:37, 4.35it/s] 15%|█▍ | 345/2338 [01:22<07:45, 4.28it/s] 15%|█▍ | 346/2338 [01:22<07:51, 4.22it/s] 15%|█▍ | 347/2338 [01:22<07:41, 4.32it/s] 15%|█▍ | 348/2338 [01:22<07:39, 4.33it/s] 15%|█▍ | 349/2338 [01:23<07:34, 4.38it/s] 15%|█▍ | 350/2338 [01:23<07:25, 4.46it/s] 15%|█▌ | 351/2338 [01:23<07:02, 4.70it/s] 15%|█▌ | 352/2338 [01:23<07:06, 4.65it/s] 15%|█▌ | 353/2338 [01:23<06:40, 4.96it/s] 15%|█▌ | 354/2338 [01:24<07:34, 4.37it/s] 15%|█▌ | 355/2338 [01:24<07:40, 4.30it/s] 15%|█▌ | 356/2338 [01:24<07:16, 4.54it/s] 15%|█▌ | 357/2338 [01:24<06:54, 4.78it/s] 15%|█▌ | 358/2338 [01:25<07:07, 4.63it/s] 15%|█▌ | 359/2338 [01:25<07:17, 4.52it/s] 15%|█▌ | 360/2338 [01:25<07:16, 4.54it/s] 15%|█▌ | 361/2338 [01:25<07:10, 4.59it/s] 15%|█▌ | 362/2338 [01:25<07:14, 4.55it/s] 16%|█▌ | 363/2338 [01:26<06:48, 4.83it/s] 16%|█▌ | 364/2338 [01:26<06:45, 4.87it/s] 16%|█▌ | 365/2338 [01:26<07:02, 4.67it/s] 16%|█▌ | 366/2338 [01:26<07:20, 4.47it/s] 16%|█▌ | 367/2338 [01:27<07:49, 4.20it/s] 16%|█▌ | 368/2338 [01:27<07:48, 4.20it/s] 16%|█▌ | 369/2338 [01:27<07:50, 4.19it/s] 16%|█▌ | 370/2338 [01:27<08:29, 3.87it/s] 16%|█▌ | 371/2338 [01:28<08:14, 3.98it/s] 16%|█▌ | 372/2338 [01:28<08:09, 4.02it/s] 16%|█▌ | 373/2338 [01:28<07:12, 4.55it/s] 16%|█▌ | 374/2338 [01:28<07:20, 4.46it/s] 16%|█▌ | 375/2338 [01:29<07:33, 4.33it/s] 16%|█▌ | 376/2338 [01:29<07:35, 4.31it/s] 16%|█▌ | 377/2338 [01:29<07:33, 4.33it/s] 16%|█▌ | 378/2338 [01:29<07:38, 4.27it/s] 16%|█▌ | 379/2338 [01:29<07:36, 4.29it/s] 16%|█▋ | 380/2338 [01:30<07:30, 4.34it/s] 16%|█▋ | 381/2338 [01:30<07:27, 4.37it/s] 16%|█▋ | 382/2338 [01:30<07:24, 4.40it/s] 16%|█▋ | 383/2338 [01:30<07:02, 4.62it/s] 16%|█▋ | 384/2338 [01:30<06:25, 5.07it/s] 16%|█▋ | 385/2338 [01:31<06:43, 4.84it/s] 17%|█▋ | 386/2338 [01:31<07:03, 4.61it/s] 17%|█▋ | 387/2338 [01:31<07:48, 4.17it/s] 17%|█▋ | 388/2338 [01:31<07:47, 4.17it/s] 17%|█▋ | 389/2338 [01:32<07:54, 4.10it/s] 17%|█▋ | 390/2338 [01:32<08:01, 4.05it/s] 17%|█▋ | 391/2338 [01:32<07:10, 4.52it/s] 17%|█▋ | 392/2338 [01:32<07:20, 4.42it/s] 17%|█▋ | 393/2338 [01:33<07:08, 4.54it/s] 17%|█▋ | 394/2338 [01:33<07:16, 4.45it/s] 17%|█▋ | 395/2338 [01:33<06:36, 4.90it/s] 17%|█▋ | 396/2338 [01:33<06:49, 4.75it/s] 17%|█▋ | 397/2338 [01:33<06:56, 4.66it/s] 17%|█▋ | 398/2338 [01:34<07:36, 4.25it/s] 17%|█▋ | 399/2338 [01:34<07:43, 4.19it/s] 17%|█▋ | 400/2338 [01:34<07:31, 4.30it/s] 17%|█▋ | 401/2338 [01:34<06:56, 4.65it/s] 17%|█▋ | 402/2338 [01:35<07:15, 4.44it/s] 17%|█▋ | 403/2338 [01:35<07:30, 4.30it/s] 17%|█▋ | 404/2338 [01:35<07:25, 4.34it/s] 17%|█▋ | 405/2338 [01:35<06:51, 4.70it/s] 17%|█▋ | 406/2338 [01:36<07:21, 4.37it/s] 17%|█▋ | 407/2338 [01:36<07:39, 4.21it/s] 17%|█▋ | 408/2338 [01:36<07:40, 4.19it/s] 17%|█▋ | 409/2338 [01:36<07:42, 4.17it/s] 18%|█▊ | 410/2338 [01:36<07:42, 4.17it/s] 18%|█▊ | 411/2338 [01:37<07:32, 4.26it/s] 18%|█▊ | 412/2338 [01:37<07:40, 4.19it/s] 18%|█▊ | 413/2338 [01:37<07:38, 4.20it/s] 18%|█▊ | 414/2338 [01:37<07:49, 4.09it/s] 18%|█▊ | 415/2338 [01:38<07:46, 4.12it/s] 18%|█▊ | 416/2338 [01:38<08:09, 3.93it/s] 18%|█▊ | 417/2338 [01:38<07:52, 4.07it/s] 18%|█▊ | 418/2338 [01:38<08:00, 3.99it/s] 18%|█▊ | 419/2338 [01:39<08:01, 3.99it/s] 18%|█▊ | 420/2338 [01:39<07:43, 4.14it/s] 18%|█▊ | 421/2338 [01:39<07:45, 4.12it/s] 18%|█▊ | 422/2338 [01:39<07:53, 4.05it/s] 18%|█▊ | 423/2338 [01:40<07:39, 4.17it/s] 18%|█▊ | 424/2338 [01:40<06:53, 4.63it/s] 18%|█▊ | 425/2338 [01:40<06:47, 4.69it/s] 18%|█▊ | 426/2338 [01:40<06:38, 4.80it/s] 18%|█▊ | 427/2338 [01:40<06:42, 4.75it/s] 18%|█▊ | 428/2338 [01:41<06:26, 4.95it/s] 18%|█▊ | 429/2338 [01:41<06:26, 4.94it/s] 18%|█▊ | 430/2338 [01:41<07:10, 4.44it/s] 18%|█▊ | 431/2338 [01:41<07:16, 4.37it/s] 18%|█▊ | 432/2338 [01:42<07:12, 4.41it/s] 19%|█▊ | 433/2338 [01:42<07:36, 4.18it/s] 19%|█▊ | 434/2338 [01:42<07:27, 4.26it/s] 19%|█▊ | 435/2338 [01:42<07:41, 4.13it/s] 19%|█▊ | 436/2338 [01:43<07:56, 3.99it/s] 19%|█▊ | 437/2338 [01:43<07:53, 4.02it/s] 19%|█▊ | 438/2338 [01:43<07:45, 4.08it/s] 19%|█▉ | 439/2338 [01:43<07:17, 4.34it/s] 19%|█▉ | 440/2338 [01:43<07:15, 4.36it/s] 19%|█▉ | 441/2338 [01:44<06:55, 4.57it/s] 19%|█▉ | 442/2338 [01:44<06:59, 4.52it/s] 19%|█▉ | 443/2338 [01:44<06:59, 4.51it/s] 19%|█▉ | 444/2338 [01:44<07:02, 4.48it/s] 19%|█▉ | 445/2338 [01:45<07:11, 4.39it/s] 19%|█▉ | 446/2338 [01:45<07:00, 4.50it/s] 19%|█▉ | 447/2338 [01:45<07:13, 4.36it/s] 19%|█▉ | 448/2338 [01:45<07:01, 4.48it/s] 19%|█▉ | 449/2338 [01:45<07:11, 4.38it/s] 19%|█▉ | 450/2338 [01:46<07:25, 4.24it/s] 19%|█▉ | 451/2338 [01:46<07:21, 4.27it/s] 19%|█▉ | 452/2338 [01:46<07:01, 4.48it/s] 19%|█▉ | 453/2338 [01:46<07:10, 4.38it/s] 19%|█▉ | 454/2338 [01:47<07:03, 4.45it/s] 19%|█▉ | 455/2338 [01:47<06:08, 5.10it/s] 20%|█▉ | 456/2338 [01:47<05:41, 5.51it/s] 20%|█▉ | 457/2338 [01:47<06:17, 4.98it/s] 20%|█▉ | 458/2338 [01:47<06:32, 4.79it/s] 20%|█▉ | 459/2338 [01:48<06:54, 4.53it/s] 20%|█▉ | 460/2338 [01:48<07:09, 4.38it/s] 20%|█▉ | 461/2338 [01:48<07:11, 4.35it/s] 20%|█▉ | 462/2338 [01:48<07:38, 4.09it/s] 20%|█▉ | 463/2338 [01:49<07:32, 4.14it/s] 20%|█▉ | 464/2338 [01:49<07:19, 4.27it/s] 20%|█▉ | 465/2338 [01:49<07:19, 4.26it/s] 20%|█▉ | 466/2338 [01:49<07:13, 4.32it/s] 20%|█▉ | 467/2338 [01:50<07:11, 4.34it/s] 20%|██ | 468/2338 [01:50<07:06, 4.39it/s] 20%|██ | 469/2338 [01:50<07:08, 4.36it/s] 20%|██ | 470/2338 [01:50<06:56, 4.49it/s] 20%|██ | 471/2338 [01:50<06:54, 4.51it/s] 20%|██ | 472/2338 [01:51<06:54, 4.50it/s] 20%|██ | 473/2338 [01:51<06:36, 4.71it/s] 20%|██ | 474/2338 [01:51<06:54, 4.49it/s] 20%|██ | 475/2338 [01:51<06:50, 4.54it/s] 20%|██ | 476/2338 [01:51<06:44, 4.60it/s] 20%|██ | 477/2338 [01:52<06:52, 4.51it/s] 20%|██ | 478/2338 [01:52<06:47, 4.57it/s] 20%|██ | 479/2338 [01:52<06:50, 4.53it/s] 21%|██ | 480/2338 [01:52<06:38, 4.66it/s] 21%|██ | 481/2338 [01:53<06:51, 4.52it/s] 21%|██ | 482/2338 [01:53<07:14, 4.27it/s] 21%|██ | 483/2338 [01:53<07:16, 4.25it/s] 21%|██ | 484/2338 [01:53<06:47, 4.55it/s] 21%|██ | 485/2338 [01:54<07:24, 4.17it/s] 21%|██ | 486/2338 [01:54<07:22, 4.18it/s] 21%|██ | 487/2338 [01:54<07:30, 4.11it/s] 21%|██ | 488/2338 [01:54<07:22, 4.18it/s] 21%|██ | 489/2338 [01:55<07:08, 4.31it/s] 21%|██ | 490/2338 [01:55<07:22, 4.17it/s] 21%|██ | 491/2338 [01:55<07:12, 4.27it/s] 21%|██ | 492/2338 [01:55<07:07, 4.32it/s] 21%|██ | 493/2338 [01:55<07:27, 4.12it/s] 21%|██ | 494/2338 [01:56<06:38, 4.63it/s] 21%|██ | 495/2338 [01:56<06:58, 4.40it/s] 21%|██ | 496/2338 [01:56<07:04, 4.34it/s] 21%|██▏ | 497/2338 [01:56<06:57, 4.41it/s] 21%|██▏ | 498/2338 [01:57<06:28, 4.74it/s] 21%|██▏ | 499/2338 [01:57<06:34, 4.66it/s] 21%|██▏ | 500/2338 [01:57<06:47, 4.51it/s] 21%|██▏ | 501/2338 [01:57<06:32, 4.68it/s] 21%|██▏ | 502/2338 [01:57<06:51, 4.46it/s] 22%|██▏ | 503/2338 [01:58<06:11, 4.94it/s] 22%|██▏ | 504/2338 [01:58<06:30, 4.70it/s] 22%|██▏ | 505/2338 [01:58<06:54, 4.42it/s] 22%|██▏ | 506/2338 [01:58<07:10, 4.25it/s] 22%|██▏ | 507/2338 [01:59<07:10, 4.25it/s] 22%|██▏ | 508/2338 [01:59<07:25, 4.11it/s] 22%|██▏ | 509/2338 [01:59<07:16, 4.19it/s] 22%|██▏ | 510/2338 [01:59<07:11, 4.24it/s] 22%|██▏ | 511/2338 [02:00<07:35, 4.01it/s] 22%|██▏ | 512/2338 [02:00<07:19, 4.15it/s] 22%|██▏ | 513/2338 [02:00<07:12, 4.22it/s] 22%|██▏ | 514/2338 [02:00<07:00, 4.34it/s] 22%|██▏ | 515/2338 [02:00<06:58, 4.36it/s] 22%|██▏ | 516/2338 [02:01<07:16, 4.18it/s] 22%|██▏ | 517/2338 [02:01<07:26, 4.08it/s] 22%|██▏ | 518/2338 [02:01<07:33, 4.01it/s] 22%|██▏ | 519/2338 [02:01<07:20, 4.13it/s] 22%|██▏ | 520/2338 [02:02<07:09, 4.23it/s] 22%|██▏ | 521/2338 [02:02<07:09, 4.23it/s] 22%|██▏ | 522/2338 [02:02<06:46, 4.46it/s] 22%|██▏ | 523/2338 [02:02<06:42, 4.51it/s] 22%|██▏ | 524/2338 [02:03<07:02, 4.29it/s] 22%|██▏ | 525/2338 [02:03<07:02, 4.29it/s] 22%|██▏ | 526/2338 [02:03<07:16, 4.15it/s] 23%|██▎ | 527/2338 [02:03<07:01, 4.30it/s] 23%|██▎ | 528/2338 [02:03<06:33, 4.61it/s] 23%|██▎ | 529/2338 [02:04<06:44, 4.48it/s] 23%|██▎ | 530/2338 [02:04<06:54, 4.36it/s] 23%|██▎ | 531/2338 [02:04<06:47, 4.44it/s] 23%|██▎ | 532/2338 [02:04<06:09, 4.89it/s] 23%|██▎ | 533/2338 [02:05<06:26, 4.67it/s] 23%|██▎ | 534/2338 [02:05<06:14, 4.81it/s] 23%|██▎ | 535/2338 [02:05<06:40, 4.50it/s] 23%|██▎ | 536/2338 [02:05<06:40, 4.50it/s] 23%|██▎ | 537/2338 [02:05<06:51, 4.38it/s] 23%|██▎ | 538/2338 [02:06<07:15, 4.13it/s] 23%|██▎ | 539/2338 [02:06<07:21, 4.07it/s] 23%|██▎ | 540/2338 [02:06<07:16, 4.12it/s] 23%|██▎ | 541/2338 [02:06<07:02, 4.25it/s] 23%|██▎ | 542/2338 [02:07<06:51, 4.36it/s] 23%|██▎ | 543/2338 [02:07<06:33, 4.56it/s] 23%|██▎ | 544/2338 [02:07<06:35, 4.53it/s] 23%|██▎ | 545/2338 [02:07<06:49, 4.38it/s] 23%|██▎ | 546/2338 [02:08<06:48, 4.39it/s] 23%|██▎ | 547/2338 [02:08<06:36, 4.52it/s] 23%|██▎ | 548/2338 [02:08<06:28, 4.60it/s] 23%|██▎ | 549/2338 [02:08<06:33, 4.54it/s] 24%|██▎ | 550/2338 [02:08<06:31, 4.57it/s] 24%|██▎ | 551/2338 [02:09<06:52, 4.33it/s] 24%|██▎ | 552/2338 [02:09<06:46, 4.39it/s] 24%|██▎ | 553/2338 [02:09<06:45, 4.40it/s] 24%|██▎ | 554/2338 [02:09<06:49, 4.36it/s] 24%|██▎ | 555/2338 [02:10<06:45, 4.40it/s] 24%|██▍ | 556/2338 [02:10<06:53, 4.31it/s] 24%|██▍ | 557/2338 [02:10<06:23, 4.64it/s] 24%|██▍ | 558/2338 [02:10<05:53, 5.04it/s] 24%|██▍ | 559/2338 [02:10<06:13, 4.76it/s] 24%|██▍ | 560/2338 [02:11<06:46, 4.38it/s] 24%|██▍ | 561/2338 [02:11<06:42, 4.42it/s] 24%|██▍ | 562/2338 [02:11<07:08, 4.14it/s] 24%|██▍ | 563/2338 [02:11<07:42, 3.84it/s] 24%|██▍ | 564/2338 [02:12<07:26, 3.97it/s] 24%|██▍ | 565/2338 [02:12<07:12, 4.10it/s] 24%|██▍ | 566/2338 [02:12<07:09, 4.12it/s] 24%|██▍ | 567/2338 [02:12<07:28, 3.95it/s] 24%|██▍ | 568/2338 [02:13<07:13, 4.08it/s] 24%|██▍ | 569/2338 [02:13<07:11, 4.10it/s] 24%|██▍ | 570/2338 [02:13<07:06, 4.14it/s] 24%|██▍ | 571/2338 [02:13<06:39, 4.42it/s] 24%|██▍ | 572/2338 [02:14<06:49, 4.31it/s] 25%|██▍ | 573/2338 [02:14<06:54, 4.26it/s] 25%|██▍ | 574/2338 [02:14<06:17, 4.68it/s] 25%|██▍ | 575/2338 [02:14<06:13, 4.72it/s] 25%|██▍ | 576/2338 [02:14<06:18, 4.65it/s] 25%|██▍ | 577/2338 [02:15<06:32, 4.48it/s] 25%|██▍ | 578/2338 [02:15<06:19, 4.63it/s] 25%|██▍ | 579/2338 [02:15<07:13, 4.06it/s] 25%|██▍ | 580/2338 [02:15<07:14, 4.04it/s] 25%|██▍ | 581/2338 [02:16<06:29, 4.51it/s] 25%|██▍ | 582/2338 [02:16<06:21, 4.60it/s] 25%|██▍ | 583/2338 [02:16<06:47, 4.30it/s] 25%|██▍ | 584/2338 [02:16<06:58, 4.20it/s] 25%|██▌ | 585/2338 [02:17<06:52, 4.25it/s] 25%|██▌ | 586/2338 [02:17<06:50, 4.26it/s] 25%|██▌ | 587/2338 [02:17<06:59, 4.17it/s] 25%|██▌ | 588/2338 [02:17<06:47, 4.29it/s] 25%|██▌ | 589/2338 [02:18<06:51, 4.25it/s] 25%|██▌ | 590/2338 [02:18<06:36, 4.41it/s] 25%|██▌ | 591/2338 [02:18<06:30, 4.48it/s] 25%|██▌ | 592/2338 [02:18<06:40, 4.36it/s] 25%|██▌ | 593/2338 [02:18<06:53, 4.22it/s] 25%|██▌ | 594/2338 [02:19<06:23, 4.54it/s] 25%|██▌ | 595/2338 [02:19<06:20, 4.58it/s] 25%|██▌ | 596/2338 [02:19<05:39, 5.14it/s] 26%|██▌ | 597/2338 [02:19<05:59, 4.84it/s] 26%|██▌ | 598/2338 [02:19<06:39, 4.36it/s] 26%|██▌ | 599/2338 [02:20<06:32, 4.43it/s] 26%|██▌ | 600/2338 [02:20<06:26, 4.50it/s] 26%|██▌ | 601/2338 [02:20<06:39, 4.35it/s] 26%|██▌ | 602/2338 [02:20<06:59, 4.14it/s] 26%|██▌ | 603/2338 [02:21<06:52, 4.21it/s] 26%|██▌ | 604/2338 [02:21<06:44, 4.29it/s] 26%|██▌ | 605/2338 [02:21<06:47, 4.25it/s] 26%|██▌ | 606/2338 [02:21<06:25, 4.49it/s] 26%|██▌ | 607/2338 [02:22<06:18, 4.58it/s] 26%|██▌ | 608/2338 [02:22<06:03, 4.76it/s] 26%|██▌ | 609/2338 [02:22<06:11, 4.65it/s] 26%|██▌ | 610/2338 [02:22<05:47, 4.98it/s] 26%|██▌ | 611/2338 [02:22<06:24, 4.49it/s] 26%|██▌ | 612/2338 [02:23<06:52, 4.19it/s] 26%|██▌ | 613/2338 [02:23<06:48, 4.22it/s] 26%|██▋ | 614/2338 [02:23<06:54, 4.16it/s] 26%|██▋ | 615/2338 [02:23<06:30, 4.41it/s] 26%|██▋ | 616/2338 [02:24<06:32, 4.39it/s] 26%|██▋ | 617/2338 [02:24<06:21, 4.51it/s] 26%|██▋ | 618/2338 [02:24<06:31, 4.39it/s] 26%|██▋ | 619/2338 [02:24<06:21, 4.51it/s] 27%|██▋ | 620/2338 [02:24<06:29, 4.41it/s] 27%|██▋ | 621/2338 [02:25<06:33, 4.37it/s] 27%|██▋ | 622/2338 [02:25<06:46, 4.23it/s] 27%|██▋ | 623/2338 [02:25<06:25, 4.44it/s] 27%|██▋ | 624/2338 [02:25<06:00, 4.75it/s] 27%|██▋ | 625/2338 [02:26<06:03, 4.72it/s] 27%|██▋ | 626/2338 [02:26<06:25, 4.44it/s] 27%|██▋ | 627/2338 [02:26<06:20, 4.50it/s] 27%|██▋ | 628/2338 [02:26<06:26, 4.42it/s] 27%|██▋ | 629/2338 [02:27<06:57, 4.09it/s] 27%|██▋ | 630/2338 [02:27<06:49, 4.17it/s] 27%|██▋ | 631/2338 [02:27<06:38, 4.28it/s] 27%|██▋ | 632/2338 [02:27<06:27, 4.41it/s] 27%|██▋ | 633/2338 [02:27<06:25, 4.43it/s] 27%|██▋ | 634/2338 [02:28<06:39, 4.27it/s] 27%|██▋ | 635/2338 [02:28<06:36, 4.30it/s] 27%|██▋ | 636/2338 [02:28<06:58, 4.06it/s] 27%|██▋ | 637/2338 [02:28<07:07, 3.98it/s] 27%|██▋ | 638/2338 [02:29<07:04, 4.01it/s] 27%|██▋ | 639/2338 [02:29<06:39, 4.25it/s] 27%|██▋ | 640/2338 [02:29<07:07, 3.97it/s] 27%|██▋ | 641/2338 [02:29<06:43, 4.20it/s] 27%|██▋ | 642/2338 [02:30<06:35, 4.29it/s] 28%|██▊ | 643/2338 [02:30<06:19, 4.47it/s] 28%|██▊ | 644/2338 [02:30<06:13, 4.53it/s] 28%|██▊ | 645/2338 [02:30<06:11, 4.56it/s] 28%|██▊ | 646/2338 [02:30<06:28, 4.35it/s] 28%|██▊ | 647/2338 [02:31<06:24, 4.40it/s] 28%|██▊ | 648/2338 [02:31<06:56, 4.06it/s] 28%|██▊ | 649/2338 [02:31<07:12, 3.91it/s] 28%|██▊ | 650/2338 [02:32<07:19, 3.84it/s] 28%|██▊ | 651/2338 [02:32<07:15, 3.87it/s] 28%|██▊ | 652/2338 [02:32<06:23, 4.39it/s] 28%|██▊ | 653/2338 [02:32<06:20, 4.42it/s] 28%|██▊ | 654/2338 [02:32<06:31, 4.30it/s] 28%|██▊ | 655/2338 [02:33<06:52, 4.08it/s] 28%|██▊ | 656/2338 [02:33<06:58, 4.01it/s] 28%|██▊ | 657/2338 [02:33<06:57, 4.03it/s] 28%|██▊ | 658/2338 [02:33<06:41, 4.18it/s] 28%|██▊ | 659/2338 [02:34<06:40, 4.19it/s] 28%|██▊ | 660/2338 [02:34<06:14, 4.48it/s] 28%|██▊ | 661/2338 [02:34<06:10, 4.52it/s] 28%|██▊ | 662/2338 [02:34<07:39, 3.65it/s] 28%|██▊ | 663/2338 [02:35<07:30, 3.72it/s] 28%|██▊ | 664/2338 [02:35<07:01, 3.97it/s] 28%|██▊ | 665/2338 [02:35<06:05, 4.58it/s] 28%|██▊ | 666/2338 [02:35<06:27, 4.31it/s] 29%|██▊ | 667/2338 [02:36<06:27, 4.32it/s] 29%|██▊ | 668/2338 [02:36<06:57, 4.00it/s] 29%|██▊ | 669/2338 [02:36<05:59, 4.64it/s] 29%|██▊ | 670/2338 [02:36<05:55, 4.70it/s] 29%|██▊ | 671/2338 [02:36<05:41, 4.88it/s] 29%|██▊ | 672/2338 [02:37<05:54, 4.70it/s] 29%|██▉ | 673/2338 [02:37<06:01, 4.61it/s] 29%|██▉ | 674/2338 [02:37<05:52, 4.72it/s] 29%|██▉ | 675/2338 [02:37<05:35, 4.95it/s] 29%|██▉ | 676/2338 [02:37<05:49, 4.75it/s] 29%|██▉ | 677/2338 [02:38<06:06, 4.53it/s] 29%|██▉ | 678/2338 [02:38<06:04, 4.55it/s] 29%|██▉ | 679/2338 [02:38<06:21, 4.35it/s] 29%|██▉ | 680/2338 [02:38<06:10, 4.47it/s] 29%|██▉ | 681/2338 [02:39<06:13, 4.44it/s] 29%|██▉ | 682/2338 [02:39<06:24, 4.31it/s] 29%|██▉ | 683/2338 [02:39<06:32, 4.22it/s] 29%|██▉ | 684/2338 [02:39<06:24, 4.31it/s] 29%|██▉ | 685/2338 [02:40<06:07, 4.50it/s] 29%|██▉ | 686/2338 [02:40<06:22, 4.32it/s] 29%|██▉ | 687/2338 [02:40<06:22, 4.31it/s] 29%|██▉ | 688/2338 [02:40<06:14, 4.40it/s] 29%|██▉ | 689/2338 [02:40<06:33, 4.19it/s] 30%|██▉ | 690/2338 [02:41<06:09, 4.46it/s] 30%|██▉ | 691/2338 [02:41<06:10, 4.45it/s] 30%|██▉ | 692/2338 [02:41<06:24, 4.28it/s] 30%|██▉ | 693/2338 [02:41<06:25, 4.26it/s] 30%|██▉ | 694/2338 [02:42<06:24, 4.28it/s] 30%|██▉ | 695/2338 [02:42<06:26, 4.25it/s] 30%|██▉ | 696/2338 [02:42<06:21, 4.30it/s] 30%|██▉ | 697/2338 [02:42<06:21, 4.30it/s] 30%|██▉ | 698/2338 [02:43<06:12, 4.40it/s] 30%|██▉ | 699/2338 [02:43<06:08, 4.45it/s] 30%|██▉ | 700/2338 [02:43<06:09, 4.43it/s] 30%|██▉ | 701/2338 [02:43<06:17, 4.34it/s] 30%|███ | 702/2338 [02:43<06:23, 4.26it/s] 30%|███ | 703/2338 [02:44<06:28, 4.20it/s] 30%|███ | 704/2338 [02:44<06:28, 4.20it/s] 30%|███ | 705/2338 [02:44<06:14, 4.36it/s] 30%|███ | 706/2338 [02:44<06:19, 4.30it/s] 30%|███ | 707/2338 [02:45<06:32, 4.16it/s] 30%|███ | 708/2338 [02:45<06:32, 4.16it/s] 30%|███ | 709/2338 [02:45<06:49, 3.98it/s] 30%|███ | 710/2338 [02:45<06:39, 4.08it/s] 30%|███ | 711/2338 [02:46<06:37, 4.09it/s] 30%|███ | 712/2338 [02:46<06:02, 4.48it/s] 30%|███ | 713/2338 [02:46<05:42, 4.74it/s] 31%|███ | 714/2338 [02:46<05:57, 4.54it/s] 31%|███ | 715/2338 [02:46<06:06, 4.43it/s] 31%|███ | 716/2338 [02:47<06:15, 4.31it/s] 31%|███ | 717/2338 [02:47<06:22, 4.24it/s] 31%|███ | 718/2338 [02:47<06:24, 4.22it/s] 31%|███ | 719/2338 [02:47<06:36, 4.08it/s] 31%|███ | 720/2338 [02:48<06:36, 4.08it/s] 31%|███ | 721/2338 [02:48<06:27, 4.18it/s] 31%|███ | 722/2338 [02:48<06:18, 4.27it/s] 31%|███ | 723/2338 [02:48<05:59, 4.50it/s] 31%|███ | 724/2338 [02:49<05:57, 4.52it/s] 31%|███ | 725/2338 [02:49<06:02, 4.45it/s] 31%|███ | 726/2338 [02:49<06:10, 4.36it/s] 31%|███ | 727/2338 [02:49<05:51, 4.59it/s] 31%|███ | 728/2338 [02:49<06:00, 4.47it/s] 31%|███ | 729/2338 [02:50<06:13, 4.31it/s] 31%|███ | 730/2338 [02:50<06:17, 4.26it/s] 31%|███▏ | 731/2338 [02:50<06:23, 4.19it/s] 31%|███▏ | 732/2338 [02:50<06:35, 4.06it/s] 31%|███▏ | 733/2338 [02:51<06:29, 4.12it/s] 31%|███▏ | 734/2338 [02:51<06:40, 4.01it/s] 31%|███▏ | 735/2338 [02:51<06:38, 4.02it/s] 31%|███▏ | 736/2338 [02:51<06:19, 4.23it/s] 32%|███▏ | 737/2338 [02:52<06:05, 4.38it/s] 32%|███▏ | 738/2338 [02:52<06:04, 4.40it/s] 32%|███▏ | 739/2338 [02:52<06:28, 4.12it/s] 32%|███▏ | 740/2338 [02:52<05:51, 4.55it/s] 32%|███▏ | 741/2338 [02:53<06:15, 4.25it/s] 32%|███▏ | 742/2338 [02:53<05:40, 4.69it/s] 32%|███▏ | 743/2338 [02:53<05:34, 4.77it/s] 32%|███▏ | 744/2338 [02:53<05:53, 4.50it/s] 32%|███▏ | 745/2338 [02:53<06:05, 4.36it/s] 32%|███▏ | 746/2338 [02:54<06:30, 4.08it/s] 32%|███▏ | 747/2338 [02:54<06:39, 3.98it/s] 32%|███▏ | 748/2338 [02:54<06:46, 3.91it/s] 32%|███▏ | 749/2338 [02:55<06:44, 3.93it/s] 32%|███▏ | 750/2338 [02:55<06:34, 4.02it/s] 32%|███▏ | 751/2338 [02:55<06:34, 4.02it/s] 32%|███▏ | 752/2338 [02:55<06:32, 4.04it/s] 32%|███▏ | 753/2338 [02:56<06:39, 3.96it/s] 32%|███▏ | 754/2338 [02:56<06:35, 4.01it/s] 32%|███▏ | 755/2338 [02:56<06:30, 4.05it/s] 32%|███▏ | 756/2338 [02:56<06:49, 3.87it/s] 32%|███▏ | 757/2338 [02:57<06:39, 3.95it/s] 32%|███▏ | 758/2338 [02:57<06:39, 3.96it/s] 32%|███▏ | 759/2338 [02:57<06:30, 4.04it/s] 33%|███▎ | 760/2338 [02:57<06:36, 3.98it/s] 33%|███▎ | 761/2338 [02:58<06:33, 4.01it/s] 33%|███▎ | 762/2338 [02:58<06:34, 3.99it/s] 33%|███▎ | 763/2338 [02:58<06:20, 4.14it/s] 33%|███▎ | 764/2338 [02:58<06:16, 4.18it/s] 33%|███▎ | 765/2338 [02:58<06:17, 4.17it/s] 33%|███▎ | 766/2338 [02:59<06:06, 4.28it/s] 33%|███▎ | 767/2338 [02:59<05:59, 4.37it/s] 33%|███▎ | 768/2338 [02:59<06:03, 4.32it/s] 33%|███▎ | 769/2338 [02:59<06:03, 4.31it/s] 33%|███▎ | 770/2338 [03:00<05:41, 4.59it/s] 33%|███▎ | 771/2338 [03:00<05:40, 4.60it/s] 33%|███▎ | 772/2338 [03:00<05:54, 4.42it/s] 33%|███▎ | 773/2338 [03:00<05:50, 4.47it/s] 33%|███▎ | 774/2338 [03:00<05:47, 4.50it/s] 33%|███▎ | 775/2338 [03:01<05:16, 4.94it/s] 33%|███▎ | 776/2338 [03:01<05:12, 5.00it/s] 33%|███▎ | 777/2338 [03:01<05:30, 4.73it/s] 33%|███▎ | 778/2338 [03:01<05:45, 4.52it/s] 33%|███▎ | 779/2338 [03:02<05:56, 4.37it/s] 33%|███▎ | 780/2338 [03:02<05:49, 4.45it/s] 33%|███▎ | 781/2338 [03:02<05:39, 4.58it/s] 33%|███▎ | 782/2338 [03:02<06:13, 4.16it/s] 33%|███▎ | 783/2338 [03:03<06:19, 4.09it/s] 34%|███▎ | 784/2338 [03:03<06:35, 3.93it/s] 34%|███▎ | 785/2338 [03:03<06:28, 4.00it/s] 34%|███▎ | 786/2338 [03:03<06:22, 4.05it/s] 34%|███▎ | 787/2338 [03:03<06:01, 4.29it/s] 34%|███▎ | 788/2338 [03:04<05:48, 4.45it/s] 34%|███▎ | 789/2338 [03:04<05:43, 4.51it/s] 34%|███▍ | 790/2338 [03:04<05:52, 4.39it/s] 34%|███▍ | 791/2338 [03:04<05:29, 4.69it/s] 34%|███▍ | 792/2338 [03:05<06:15, 4.12it/s] 34%|███▍ | 793/2338 [03:05<06:11, 4.16it/s] 34%|███▍ | 794/2338 [03:05<06:24, 4.01it/s] 34%|███▍ | 795/2338 [03:05<06:12, 4.15it/s] 34%|███▍ | 796/2338 [03:06<06:24, 4.01it/s] 34%|███▍ | 797/2338 [03:06<06:21, 4.04it/s] 34%|███▍ | 798/2338 [03:06<06:17, 4.08it/s] 34%|███▍ | 799/2338 [03:06<06:19, 4.06it/s] 34%|███▍ | 800/2338 [03:07<05:45, 4.45it/s] 34%|███▍ | 801/2338 [03:07<05:40, 4.51it/s] 34%|███▍ | 802/2338 [03:07<05:27, 4.69it/s] 34%|███▍ | 803/2338 [03:07<05:24, 4.73it/s] 34%|███▍ | 804/2338 [03:07<05:33, 4.61it/s] 34%|███▍ | 805/2338 [03:08<05:32, 4.60it/s] 34%|███▍ | 806/2338 [03:08<05:56, 4.30it/s] 35%|███▍ | 807/2338 [03:08<05:19, 4.79it/s] 35%|███▍ | 808/2338 [03:08<05:28, 4.65it/s] 35%|███▍ | 809/2338 [03:08<05:44, 4.44it/s] 35%|███▍ | 810/2338 [03:09<06:04, 4.20it/s] 35%|███▍ | 811/2338 [03:09<05:52, 4.33it/s] 35%|███▍ | 812/2338 [03:09<06:05, 4.18it/s] 35%|███▍ | 813/2338 [03:09<05:52, 4.33it/s] 35%|███▍ | 814/2338 [03:10<05:38, 4.50it/s] 35%|███▍ | 815/2338 [03:10<05:30, 4.61it/s] 35%|███▍ | 816/2338 [03:10<05:49, 4.36it/s] 35%|███▍ | 817/2338 [03:10<06:05, 4.16it/s] 35%|███▍ | 818/2338 [03:11<06:08, 4.13it/s] 35%|███▌ | 819/2338 [03:11<06:20, 3.99it/s] 35%|███▌ | 820/2338 [03:11<06:18, 4.01it/s] 35%|███▌ | 821/2338 [03:11<06:16, 4.03it/s] 35%|███▌ | 822/2338 [03:12<06:06, 4.13it/s] 35%|███▌ | 823/2338 [03:12<06:23, 3.95it/s] 35%|███▌ | 824/2338 [03:12<06:11, 4.07it/s] 35%|███▌ | 825/2338 [03:12<06:02, 4.18it/s] 35%|███▌ | 826/2338 [03:13<06:06, 4.13it/s] 35%|███▌ | 827/2338 [03:13<05:59, 4.21it/s] 35%|███▌ | 828/2338 [03:13<05:44, 4.38it/s] 35%|███▌ | 829/2338 [03:13<05:50, 4.30it/s] 36%|███▌ | 830/2338 [03:13<05:49, 4.31it/s] 36%|███▌ | 831/2338 [03:14<05:50, 4.31it/s] 36%|███▌ | 832/2338 [03:14<05:30, 4.55it/s] 36%|███▌ | 833/2338 [03:14<05:24, 4.63it/s] 36%|███▌ | 834/2338 [03:14<05:28, 4.58it/s] 36%|███▌ | 835/2338 [03:15<05:24, 4.63it/s] 36%|███▌ | 836/2338 [03:15<05:34, 4.49it/s] 36%|███▌ | 837/2338 [03:15<05:49, 4.30it/s] 36%|███▌ | 838/2338 [03:15<05:39, 4.42it/s] 36%|███▌ | 839/2338 [03:15<05:32, 4.50it/s] 36%|███▌ | 840/2338 [03:16<05:32, 4.51it/s] 36%|███▌ | 841/2338 [03:16<05:23, 4.62it/s] 36%|███▌ | 842/2338 [03:16<05:42, 4.36it/s] 36%|███▌ | 843/2338 [03:16<05:53, 4.23it/s] 36%|███▌ | 844/2338 [03:17<05:48, 4.29it/s] 36%|███▌ | 845/2338 [03:17<05:56, 4.19it/s] 36%|███▌ | 846/2338 [03:17<05:49, 4.27it/s] 36%|███▌ | 847/2338 [03:17<05:23, 4.62it/s] 36%|███▋ | 848/2338 [03:17<05:16, 4.71it/s] 36%|███▋ | 849/2338 [03:18<05:03, 4.91it/s] 36%|███▋ | 850/2338 [03:18<05:25, 4.57it/s] 36%|███▋ | 851/2338 [03:18<05:23, 4.60it/s] 36%|███▋ | 852/2338 [03:18<05:29, 4.50it/s] 36%|███▋ | 853/2338 [03:19<05:43, 4.32it/s] 37%|███▋ | 854/2338 [03:19<05:42, 4.33it/s] 37%|███▋ | 855/2338 [03:19<06:01, 4.10it/s] 37%|███▋ | 856/2338 [03:19<05:57, 4.15it/s] 37%|███▋ | 857/2338 [03:20<05:51, 4.21it/s] 37%|███▋ | 858/2338 [03:20<06:07, 4.02it/s] 37%|███▋ | 859/2338 [03:20<05:56, 4.15it/s] 37%|███▋ | 860/2338 [03:20<06:03, 4.07it/s] 37%|███▋ | 861/2338 [03:21<05:46, 4.26it/s] 37%|███▋ | 862/2338 [03:21<05:39, 4.35it/s] 37%|███▋ | 863/2338 [03:21<05:45, 4.26it/s] 37%|███▋ | 864/2338 [03:21<05:48, 4.23it/s] 37%|███▋ | 865/2338 [03:21<05:44, 4.27it/s] 37%|███▋ | 866/2338 [03:22<05:48, 4.23it/s] 37%|███▋ | 867/2338 [03:22<05:52, 4.18it/s] 37%|███▋ | 868/2338 [03:22<05:42, 4.30it/s] 37%|███▋ | 869/2338 [03:22<05:30, 4.44it/s] 37%|███▋ | 870/2338 [03:23<05:30, 4.44it/s] 37%|███▋ | 871/2338 [03:23<05:34, 4.38it/s] 37%|███▋ | 872/2338 [03:23<06:00, 4.07it/s] 37%|███▋ | 873/2338 [03:23<05:58, 4.09it/s] 37%|███▋ | 874/2338 [03:24<05:51, 4.16it/s] 37%|███▋ | 875/2338 [03:24<06:00, 4.06it/s] 37%|███▋ | 876/2338 [03:24<05:54, 4.12it/s] 38%|███▊ | 877/2338 [03:24<06:15, 3.89it/s] 38%|███▊ | 878/2338 [03:25<06:02, 4.03it/s] 38%|███▊ | 879/2338 [03:25<07:36, 3.20it/s] 38%|███▊ | 880/2338 [03:25<06:56, 3.50it/s] 38%|███▊ | 881/2338 [03:26<06:32, 3.71it/s] 38%|███▊ | 882/2338 [03:26<06:21, 3.82it/s] 38%|███▊ | 883/2338 [03:26<06:12, 3.91it/s] 38%|███▊ | 884/2338 [03:26<05:59, 4.05it/s] 38%|███▊ | 885/2338 [03:26<05:48, 4.17it/s] 38%|███▊ | 886/2338 [03:27<05:48, 4.17it/s] 38%|███▊ | 887/2338 [03:27<05:37, 4.30it/s] 38%|███▊ | 888/2338 [03:27<05:22, 4.50it/s] 38%|███▊ | 889/2338 [03:27<05:31, 4.37it/s] 38%|███▊ | 890/2338 [03:28<05:06, 4.72it/s] 38%|███▊ | 891/2338 [03:28<05:21, 4.50it/s] 38%|███▊ | 892/2338 [03:28<05:38, 4.28it/s] 38%|███▊ | 893/2338 [03:28<05:12, 4.63it/s] 38%|███▊ | 894/2338 [03:28<05:26, 4.42it/s] 38%|███▊ | 895/2338 [03:29<05:42, 4.21it/s] 38%|███▊ | 896/2338 [03:29<05:50, 4.11it/s] 38%|███▊ | 897/2338 [03:29<05:46, 4.15it/s] 38%|███▊ | 898/2338 [03:29<05:44, 4.18it/s] 38%|███▊ | 899/2338 [03:30<05:53, 4.07it/s] 38%|███▊ | 900/2338 [03:30<05:54, 4.05it/s] 39%|███▊ | 901/2338 [03:30<05:57, 4.02it/s] 39%|███▊ | 902/2338 [03:30<05:44, 4.17it/s] 39%|███▊ | 903/2338 [03:31<05:51, 4.08it/s] 39%|███▊ | 904/2338 [03:31<05:51, 4.08it/s] 39%|███▊ | 905/2338 [03:31<05:44, 4.16it/s] 39%|███▉ | 906/2338 [03:31<05:52, 4.06it/s] 39%|███▉ | 907/2338 [03:32<05:53, 4.05it/s] 39%|███▉ | 908/2338 [03:32<05:55, 4.03it/s] 39%|███▉ | 909/2338 [03:32<05:53, 4.05it/s] 39%|███▉ | 910/2338 [03:32<05:51, 4.06it/s] 39%|███▉ | 911/2338 [03:33<05:45, 4.12it/s] 39%|███▉ | 912/2338 [03:33<05:46, 4.11it/s] 39%|███▉ | 913/2338 [03:33<05:46, 4.12it/s] 39%|███▉ | 914/2338 [03:33<05:43, 4.14it/s] 39%|███▉ | 915/2338 [03:34<05:48, 4.09it/s] 39%|███▉ | 916/2338 [03:34<05:39, 4.19it/s] 39%|███▉ | 917/2338 [03:34<05:49, 4.07it/s] 39%|███▉ | 918/2338 [03:34<05:48, 4.08it/s] 39%|███▉ | 919/2338 [03:35<06:00, 3.93it/s] 39%|███▉ | 920/2338 [03:35<05:36, 4.22it/s] 39%|███▉ | 921/2338 [03:35<05:39, 4.18it/s] 39%|███▉ | 922/2338 [03:35<05:11, 4.55it/s] 39%|███▉ | 923/2338 [03:36<05:18, 4.45it/s] 40%|███▉ | 924/2338 [03:36<05:32, 4.25it/s] 40%|███▉ | 925/2338 [03:36<05:28, 4.30it/s] 40%|███▉ | 926/2338 [03:36<05:30, 4.28it/s] 40%|███▉ | 927/2338 [03:36<05:34, 4.22it/s] 40%|███▉ | 928/2338 [03:37<05:25, 4.33it/s] 40%|███▉ | 929/2338 [03:37<05:29, 4.28it/s] 40%|███▉ | 930/2338 [03:37<05:27, 4.30it/s] 40%|███▉ | 931/2338 [03:37<04:54, 4.77it/s] 40%|███▉ | 932/2338 [03:38<05:05, 4.61it/s] 40%|███▉ | 933/2338 [03:38<05:19, 4.39it/s] 40%|███▉ | 934/2338 [03:38<05:17, 4.43it/s] 40%|███▉ | 935/2338 [03:38<05:02, 4.63it/s] 40%|████ | 936/2338 [03:38<05:04, 4.60it/s] 40%|████ | 937/2338 [03:39<05:32, 4.21it/s] 40%|████ | 938/2338 [03:39<05:03, 4.62it/s] 40%|████ | 939/2338 [03:39<05:04, 4.59it/s] 40%|████ | 940/2338 [03:39<05:06, 4.56it/s] 40%|████ | 941/2338 [03:40<05:32, 4.20it/s] 40%|████ | 942/2338 [03:40<05:34, 4.17it/s] 40%|████ | 943/2338 [03:40<05:25, 4.28it/s] 40%|████ | 944/2338 [03:40<05:14, 4.43it/s] 40%|████ | 945/2338 [03:41<05:22, 4.31it/s] 40%|████ | 946/2338 [03:41<05:22, 4.31it/s] 41%|████ | 947/2338 [03:41<05:22, 4.31it/s] 41%|████ | 948/2338 [03:41<05:13, 4.43it/s] 41%|████ | 949/2338 [03:41<05:30, 4.20it/s] 41%|████ | 950/2338 [03:42<05:27, 4.24it/s] 41%|████ | 951/2338 [03:42<05:31, 4.18it/s] 41%|████ | 952/2338 [03:42<05:29, 4.21it/s] 41%|████ | 953/2338 [03:42<05:34, 4.15it/s] 41%|████ | 954/2338 [03:43<05:34, 4.14it/s] 41%|████ | 955/2338 [03:43<05:33, 4.15it/s] 41%|████ | 956/2338 [03:43<05:38, 4.08it/s] 41%|████ | 957/2338 [03:43<05:39, 4.06it/s] 41%|████ | 958/2338 [03:44<05:37, 4.09it/s] 41%|████ | 959/2338 [03:44<05:44, 4.00it/s] 41%|████ | 960/2338 [03:44<05:44, 4.00it/s] 41%|████ | 961/2338 [03:44<05:26, 4.21it/s] 41%|████ | 962/2338 [03:45<05:22, 4.27it/s] 41%|████ | 963/2338 [03:45<05:24, 4.24it/s] 41%|████ | 964/2338 [03:45<05:45, 3.98it/s] 41%|████▏ | 965/2338 [03:45<05:23, 4.24it/s] 41%|████▏ | 966/2338 [03:46<04:54, 4.66it/s] 41%|████▏ | 967/2338 [03:46<05:08, 4.45it/s] 41%|████▏ | 968/2338 [03:46<05:15, 4.34it/s] 41%|████▏ | 969/2338 [03:46<05:20, 4.27it/s] 41%|████▏ | 970/2338 [03:46<05:19, 4.28it/s] 42%|████▏ | 971/2338 [03:47<05:18, 4.30it/s] 42%|████▏ | 972/2338 [03:47<05:19, 4.27it/s] 42%|████▏ | 973/2338 [03:47<05:21, 4.25it/s] 42%|████▏ | 974/2338 [03:47<05:17, 4.29it/s] 42%|████▏ | 975/2338 [03:48<05:23, 4.21it/s] 42%|████▏ | 976/2338 [03:48<05:13, 4.35it/s] 42%|████▏ | 977/2338 [03:48<04:57, 4.58it/s] 42%|████▏ | 978/2338 [03:48<05:26, 4.17it/s] 42%|████▏ | 979/2338 [03:49<05:05, 4.44it/s] 42%|████▏ | 980/2338 [03:49<05:04, 4.46it/s] 42%|████▏ | 981/2338 [03:49<05:08, 4.39it/s] 42%|████▏ | 982/2338 [03:49<05:11, 4.36it/s] 42%|████▏ | 983/2338 [03:49<05:24, 4.17it/s] 42%|████▏ | 984/2338 [03:50<04:58, 4.53it/s] 42%|████▏ | 985/2338 [03:50<05:01, 4.48it/s] 42%|████▏ | 986/2338 [03:50<05:06, 4.41it/s] 42%|████▏ | 987/2338 [03:50<04:45, 4.73it/s] 42%|████▏ | 988/2338 [03:51<04:53, 4.61it/s] 42%|████▏ | 989/2338 [03:51<05:06, 4.41it/s] 42%|████▏ | 990/2338 [03:51<05:06, 4.40it/s] 42%|████▏ | 991/2338 [03:51<04:47, 4.68it/s] 42%|████▏ | 992/2338 [03:51<04:27, 5.02it/s] 42%|████▏ | 993/2338 [03:52<04:35, 4.88it/s] 43%|████▎ | 994/2338 [03:52<04:55, 4.55it/s] 43%|████▎ | 995/2338 [03:52<04:59, 4.49it/s] 43%|████▎ | 996/2338 [03:52<05:02, 4.44it/s] 43%|████▎ | 997/2338 [03:53<05:05, 4.38it/s] 43%|████▎ | 998/2338 [03:53<05:03, 4.42it/s] 43%|████▎ | 999/2338 [03:53<05:08, 4.34it/s] 43%|████▎ | 1000/2338 [03:53<05:11, 4.29it/s] 43%|████▎ | 1001/2338 [03:53<04:43, 4.72it/s] 43%|████▎ | 1002/2338 [03:54<04:45, 4.68it/s] 43%|████▎ | 1003/2338 [03:54<04:57, 4.49it/s] 43%|████▎ | 1004/2338 [03:54<05:14, 4.24it/s] 43%|████▎ | 1005/2338 [03:54<05:15, 4.23it/s] 43%|████▎ | 1006/2338 [03:55<05:05, 4.36it/s] 43%|████▎ | 1007/2338 [03:55<05:12, 4.26it/s] 43%|████▎ | 1008/2338 [03:55<05:13, 4.24it/s] 43%|████▎ | 1009/2338 [03:55<05:30, 4.03it/s] 43%|████▎ | 1010/2338 [03:56<05:08, 4.31it/s] 43%|████▎ | 1011/2338 [03:56<04:39, 4.74it/s] 43%|████▎ | 1012/2338 [03:56<04:16, 5.18it/s] 43%|████▎ | 1013/2338 [03:56<04:20, 5.08it/s] 43%|████▎ | 1014/2338 [03:56<04:33, 4.84it/s] 43%|████▎ | 1015/2338 [03:57<04:56, 4.46it/s] 43%|████▎ | 1016/2338 [03:57<04:57, 4.44it/s] 43%|████▎ | 1017/2338 [03:57<04:43, 4.66it/s] 44%|████▎ | 1018/2338 [03:57<04:34, 4.80it/s] 44%|████▎ | 1019/2338 [03:57<04:50, 4.54it/s] 44%|████▎ | 1020/2338 [03:58<04:47, 4.59it/s] 44%|████▎ | 1021/2338 [03:58<04:54, 4.46it/s] 44%|████▎ | 1022/2338 [03:58<04:54, 4.48it/s] 44%|████▍ | 1023/2338 [03:58<04:57, 4.42it/s] 44%|████▍ | 1024/2338 [03:58<04:28, 4.89it/s] 44%|████▍ | 1025/2338 [03:59<05:02, 4.34it/s] 44%|████▍ | 1026/2338 [03:59<04:53, 4.47it/s] 44%|████▍ | 1027/2338 [03:59<04:48, 4.54it/s] 44%|████▍ | 1028/2338 [03:59<04:52, 4.48it/s] 44%|████▍ | 1029/2338 [04:00<04:41, 4.66it/s] 44%|████▍ | 1030/2338 [04:00<04:35, 4.74it/s] 44%|████▍ | 1031/2338 [04:00<04:22, 4.97it/s] 44%|████▍ | 1032/2338 [04:00<04:35, 4.74it/s] 44%|████▍ | 1033/2338 [04:00<04:37, 4.69it/s] 44%|████▍ | 1034/2338 [04:01<04:42, 4.62it/s] 44%|████▍ | 1035/2338 [04:01<04:43, 4.60it/s] 44%|████▍ | 1036/2338 [04:01<04:50, 4.48it/s] 44%|████▍ | 1037/2338 [04:01<05:02, 4.30it/s] 44%|████▍ | 1038/2338 [04:02<05:02, 4.30it/s] 44%|████▍ | 1039/2338 [04:02<04:38, 4.67it/s] 44%|████▍ | 1040/2338 [04:02<04:34, 4.72it/s] 45%|████▍ | 1041/2338 [04:02<04:41, 4.61it/s] 45%|████▍ | 1042/2338 [04:02<04:53, 4.42it/s] 45%|████▍ | 1043/2338 [04:03<04:51, 4.45it/s] 45%|████▍ | 1044/2338 [04:03<04:49, 4.47it/s] 45%|████▍ | 1045/2338 [04:03<04:55, 4.37it/s] 45%|████▍ | 1046/2338 [04:03<05:10, 4.16it/s] 45%|████▍ | 1047/2338 [04:04<05:24, 3.98it/s] 45%|████▍ | 1048/2338 [04:04<05:23, 3.98it/s] 45%|████▍ | 1049/2338 [04:04<05:18, 4.04it/s] 45%|████▍ | 1050/2338 [04:04<05:14, 4.10it/s] 45%|████▍ | 1051/2338 [04:05<05:06, 4.20it/s] 45%|████▍ | 1052/2338 [04:05<05:08, 4.17it/s] 45%|████▌ | 1053/2338 [04:05<05:13, 4.10it/s] 45%|████▌ | 1054/2338 [04:05<05:12, 4.11it/s] 45%|████▌ | 1055/2338 [04:06<05:08, 4.16it/s] 45%|████▌ | 1056/2338 [04:06<05:12, 4.10it/s] 45%|████▌ | 1057/2338 [04:06<05:09, 4.14it/s] 45%|████▌ | 1058/2338 [04:06<05:08, 4.16it/s] 45%|████▌ | 1059/2338 [04:07<05:16, 4.04it/s] 45%|████▌ | 1060/2338 [04:07<05:08, 4.15it/s] 45%|████▌ | 1061/2338 [04:07<05:05, 4.18it/s] 45%|████▌ | 1062/2338 [04:07<05:19, 3.99it/s] 45%|████▌ | 1063/2338 [04:08<05:21, 3.96it/s] 46%|████▌ | 1064/2338 [04:08<05:14, 4.06it/s] 46%|████▌ | 1065/2338 [04:08<05:09, 4.11it/s] 46%|████▌ | 1066/2338 [04:08<05:03, 4.19it/s] 46%|████▌ | 1067/2338 [04:08<04:46, 4.44it/s] 46%|████▌ | 1068/2338 [04:09<04:47, 4.42it/s] 46%|████▌ | 1069/2338 [04:09<04:51, 4.35it/s] 46%|████▌ | 1070/2338 [04:09<04:42, 4.48it/s] 46%|████▌ | 1071/2338 [04:09<04:29, 4.69it/s] 46%|████▌ | 1072/2338 [04:10<04:25, 4.76it/s] 46%|████▌ | 1073/2338 [04:10<04:23, 4.81it/s] 46%|████▌ | 1074/2338 [04:10<04:32, 4.65it/s] 46%|████▌ | 1075/2338 [04:10<04:30, 4.68it/s] 46%|████▌ | 1076/2338 [04:10<04:28, 4.69it/s] 46%|████▌ | 1077/2338 [04:11<04:36, 4.56it/s] 46%|████▌ | 1078/2338 [04:11<04:51, 4.32it/s] 46%|████▌ | 1079/2338 [04:11<04:57, 4.23it/s] 46%|████▌ | 1080/2338 [04:11<04:50, 4.33it/s] 46%|████▌ | 1081/2338 [04:12<04:34, 4.58it/s] 46%|████▋ | 1082/2338 [04:12<04:27, 4.69it/s] 46%|████▋ | 1083/2338 [04:12<04:35, 4.55it/s] 46%|████▋ | 1084/2338 [04:12<04:53, 4.27it/s] 46%|████▋ | 1085/2338 [04:12<04:52, 4.29it/s] 46%|████▋ | 1086/2338 [04:13<05:02, 4.13it/s] 46%|████▋ | 1087/2338 [04:13<05:02, 4.13it/s] 47%|████▋ | 1088/2338 [04:13<05:10, 4.03it/s] 47%|████▋ | 1089/2338 [04:14<05:11, 4.01it/s] 47%|████▋ | 1090/2338 [04:14<05:04, 4.10it/s] 47%|████▋ | 1091/2338 [04:14<05:05, 4.08it/s] 47%|████▋ | 1092/2338 [04:14<05:04, 4.10it/s] 47%|████▋ | 1093/2338 [04:14<04:52, 4.26it/s] 47%|████▋ | 1094/2338 [04:15<04:53, 4.24it/s] 47%|████▋ | 1095/2338 [04:15<04:56, 4.19it/s] 47%|████▋ | 1096/2338 [04:15<05:12, 3.97it/s] 47%|████▋ | 1097/2338 [04:15<05:01, 4.11it/s] 47%|████▋ | 1098/2338 [04:16<04:57, 4.17it/s] 47%|████▋ | 1099/2338 [04:16<04:56, 4.18it/s] 47%|████▋ | 1100/2338 [04:16<04:59, 4.14it/s] 47%|████▋ | 1101/2338 [04:16<05:10, 3.98it/s] 47%|████▋ | 1102/2338 [04:17<04:57, 4.15it/s] 47%|████▋ | 1103/2338 [04:17<04:58, 4.14it/s] 47%|████▋ | 1104/2338 [04:17<05:15, 3.91it/s] 47%|████▋ | 1105/2338 [04:17<05:09, 3.98it/s] 47%|████▋ | 1106/2338 [04:18<04:56, 4.15it/s] 47%|████▋ | 1107/2338 [04:18<05:09, 3.97it/s] 47%|████▋ | 1108/2338 [04:18<05:14, 3.91it/s] 47%|████▋ | 1109/2338 [04:18<05:17, 3.87it/s] 47%|████▋ | 1110/2338 [04:19<05:11, 3.94it/s] 48%|████▊ | 1111/2338 [04:19<04:45, 4.29it/s] 48%|████▊ | 1112/2338 [04:19<04:39, 4.38it/s] 48%|████▊ | 1113/2338 [04:19<04:33, 4.49it/s] 48%|████▊ | 1114/2338 [04:20<04:40, 4.37it/s] 48%|████▊ | 1115/2338 [04:20<04:17, 4.76it/s] 48%|████▊ | 1116/2338 [04:20<04:29, 4.53it/s] 48%|████▊ | 1117/2338 [04:20<04:24, 4.62it/s] 48%|████▊ | 1118/2338 [04:20<04:20, 4.68it/s] 48%|████▊ | 1119/2338 [04:21<04:35, 4.42it/s] 48%|████▊ | 1120/2338 [04:21<04:39, 4.35it/s] 48%|████▊ | 1121/2338 [04:21<04:28, 4.54it/s] 48%|████▊ | 1122/2338 [04:21<04:32, 4.46it/s] 48%|████▊ | 1123/2338 [04:22<04:32, 4.45it/s] 48%|████▊ | 1124/2338 [04:22<04:40, 4.32it/s] 48%|████▊ | 1125/2338 [04:22<05:02, 4.01it/s] 48%|████▊ | 1126/2338 [04:22<04:51, 4.16it/s] 48%|████▊ | 1127/2338 [04:23<04:53, 4.12it/s] 48%|████▊ | 1128/2338 [04:23<05:05, 3.96it/s] 48%|████▊ | 1129/2338 [04:23<05:02, 3.99it/s] 48%|████▊ | 1130/2338 [04:23<04:53, 4.12it/s] 48%|████▊ | 1131/2338 [04:24<04:50, 4.16it/s] 48%|████▊ | 1132/2338 [04:24<04:59, 4.03it/s] 48%|████▊ | 1133/2338 [04:24<05:01, 4.00it/s] 49%|████▊ | 1134/2338 [04:24<04:41, 4.28it/s] 49%|████▊ | 1135/2338 [04:24<04:41, 4.28it/s] 49%|████▊ | 1136/2338 [04:25<04:45, 4.21it/s] 49%|████▊ | 1137/2338 [04:25<04:41, 4.26it/s] 49%|████▊ | 1138/2338 [04:25<04:39, 4.29it/s] 49%|████▊ | 1139/2338 [04:25<04:36, 4.34it/s] 49%|████▉ | 1140/2338 [04:26<04:34, 4.37it/s] 49%|████▉ | 1141/2338 [04:26<04:33, 4.38it/s] 49%|████▉ | 1142/2338 [04:26<04:31, 4.40it/s] 49%|████▉ | 1143/2338 [04:26<04:26, 4.49it/s] 49%|████▉ | 1144/2338 [04:26<04:24, 4.51it/s] 49%|████▉ | 1145/2338 [04:27<04:17, 4.64it/s] 49%|████▉ | 1146/2338 [04:27<04:23, 4.53it/s] 49%|████▉ | 1147/2338 [04:27<04:32, 4.37it/s] 49%|████▉ | 1148/2338 [04:27<04:30, 4.41it/s] 49%|████▉ | 1149/2338 [04:28<04:21, 4.55it/s] 49%|████▉ | 1150/2338 [04:28<04:26, 4.47it/s] 49%|████▉ | 1151/2338 [04:28<04:31, 4.36it/s] 49%|████▉ | 1152/2338 [04:28<04:07, 4.78it/s] 49%|████▉ | 1153/2338 [04:28<04:03, 4.87it/s] 49%|████▉ | 1154/2338 [04:29<04:30, 4.38it/s] 49%|████▉ | 1155/2338 [04:29<04:37, 4.26it/s] 49%|████▉ | 1156/2338 [04:29<04:40, 4.22it/s] 49%|████▉ | 1157/2338 [04:29<04:45, 4.14it/s] 50%|████▉ | 1158/2338 [04:30<04:36, 4.27it/s] 50%|████▉ | 1159/2338 [04:30<04:41, 4.19it/s] 50%|████▉ | 1160/2338 [04:30<04:33, 4.31it/s] 50%|████▉ | 1161/2338 [04:30<04:25, 4.43it/s] 50%|████▉ | 1162/2338 [04:31<04:11, 4.69it/s] 50%|████▉ | 1163/2338 [04:31<04:18, 4.54it/s] 50%|████▉ | 1164/2338 [04:31<04:37, 4.23it/s] 50%|████▉ | 1165/2338 [04:31<04:07, 4.74it/s] 50%|████▉ | 1166/2338 [04:31<04:20, 4.50it/s] 50%|████▉ | 1167/2338 [04:32<04:46, 4.09it/s] 50%|████▉ | 1168/2338 [04:32<04:37, 4.22it/s] 50%|█████ | 1169/2338 [04:32<04:36, 4.22it/s] 50%|█████ | 1170/2338 [04:32<04:30, 4.32it/s] 50%|█████ | 1171/2338 [04:33<04:44, 4.10it/s] 50%|█████ | 1172/2338 [04:33<04:19, 4.49it/s] 50%|█████ | 1173/2338 [04:33<04:31, 4.29it/s] 50%|█████ | 1174/2338 [04:33<04:26, 4.36it/s] 50%|█████ | 1175/2338 [04:34<04:29, 4.32it/s] 50%|█████ | 1176/2338 [04:34<04:32, 4.26it/s] 50%|█████ | 1177/2338 [04:34<04:41, 4.12it/s] 50%|█████ | 1178/2338 [04:34<04:38, 4.17it/s] 50%|█████ | 1179/2338 [04:35<04:48, 4.01it/s] 50%|█████ | 1180/2338 [04:35<04:45, 4.06it/s] 51%|█████ | 1181/2338 [04:35<04:57, 3.89it/s] 51%|█████ | 1182/2338 [04:35<04:46, 4.03it/s] 51%|█████ | 1183/2338 [04:36<04:45, 4.05it/s] 51%|█████ | 1184/2338 [04:36<04:45, 4.04it/s] 51%|█████ | 1185/2338 [04:36<04:41, 4.09it/s] 51%|█████ | 1186/2338 [04:36<04:35, 4.18it/s] 51%|█████ | 1187/2338 [04:37<04:34, 4.20it/s] 51%|█████ | 1188/2338 [04:37<04:30, 4.25it/s] 51%|█████ | 1189/2338 [04:37<04:34, 4.19it/s] 51%|█████ | 1190/2338 [04:37<04:35, 4.16it/s] 51%|█████ | 1191/2338 [04:37<04:35, 4.17it/s] 51%|█████ | 1192/2338 [04:38<03:59, 4.79it/s] 51%|█████ | 1193/2338 [04:38<04:13, 4.52it/s] 51%|█████ | 1194/2338 [04:38<04:14, 4.49it/s] 51%|█████ | 1195/2338 [04:38<04:21, 4.37it/s] 51%|█████ | 1196/2338 [04:38<03:53, 4.88it/s] 51%|█████ | 1197/2338 [04:39<03:57, 4.80it/s] 51%|█████ | 1198/2338 [04:39<04:06, 4.62it/s] 51%|█████▏ | 1199/2338 [04:39<04:08, 4.58it/s] 51%|█████▏ | 1200/2338 [04:39<04:22, 4.34it/s] 51%|█████▏ | 1201/2338 [04:40<04:21, 4.35it/s] 51%|█████▏ | 1202/2338 [04:40<04:29, 4.22it/s] 51%|█████▏ | 1203/2338 [04:40<04:34, 4.14it/s] 51%|█████▏ | 1204/2338 [04:40<04:41, 4.03it/s] 52%|█████▏ | 1205/2338 [04:41<04:42, 4.02it/s] 52%|█████▏ | 1206/2338 [04:41<04:46, 3.96it/s] 52%|█████▏ | 1207/2338 [04:41<04:45, 3.97it/s] 52%|█████▏ | 1208/2338 [04:41<04:34, 4.11it/s] 52%|█████▏ | 1209/2338 [04:42<04:43, 3.98it/s] 52%|█████▏ | 1210/2338 [04:42<04:49, 3.90it/s] 52%|█████▏ | 1211/2338 [04:42<04:41, 4.00it/s] 52%|█████▏ | 1212/2338 [04:42<04:40, 4.01it/s] 52%|█████▏ | 1213/2338 [04:43<04:33, 4.11it/s] 52%|█████▏ | 1214/2338 [04:43<04:21, 4.29it/s] 52%|█████▏ | 1215/2338 [04:43<04:18, 4.34it/s] 52%|█████▏ | 1216/2338 [04:43<04:24, 4.25it/s] 52%|█████▏ | 1217/2338 [04:44<04:34, 4.09it/s] 52%|█████▏ | 1218/2338 [04:44<04:20, 4.29it/s] 52%|█████▏ | 1219/2338 [04:44<04:28, 4.17it/s] 52%|█████▏ | 1220/2338 [04:44<04:40, 3.99it/s] 52%|█████▏ | 1221/2338 [04:45<04:24, 4.22it/s] 52%|█████▏ | 1222/2338 [04:45<04:21, 4.27it/s] 52%|█████▏ | 1223/2338 [04:45<04:32, 4.09it/s] 52%|█████▏ | 1224/2338 [04:45<04:33, 4.07it/s] 52%|█████▏ | 1225/2338 [04:46<04:46, 3.88it/s] 52%|█████▏ | 1226/2338 [04:46<04:41, 3.95it/s] 52%|█████▏ | 1227/2338 [04:46<04:37, 4.00it/s] 53%|█████▎ | 1228/2338 [04:46<04:32, 4.08it/s] 53%|█████▎ | 1229/2338 [04:47<04:20, 4.26it/s] 53%|█████▎ | 1230/2338 [04:47<04:30, 4.09it/s] 53%|█████▎ | 1231/2338 [04:47<04:28, 4.12it/s] 53%|█████▎ | 1232/2338 [04:47<04:27, 4.14it/s] 53%|█████▎ | 1233/2338 [04:48<04:29, 4.10it/s] 53%|█████▎ | 1234/2338 [04:48<04:20, 4.25it/s] 53%|█████▎ | 1235/2338 [04:48<04:20, 4.23it/s] 53%|█████▎ | 1236/2338 [04:48<04:26, 4.13it/s] 53%|█████▎ | 1237/2338 [04:48<04:34, 4.02it/s] 53%|█████▎ | 1238/2338 [04:49<04:29, 4.09it/s] 53%|█████▎ | 1239/2338 [04:49<04:58, 3.69it/s] 53%|█████▎ | 1240/2338 [04:49<04:56, 3.70it/s] 53%|█████▎ | 1241/2338 [04:50<04:42, 3.89it/s] 53%|█████▎ | 1242/2338 [04:50<04:06, 4.45it/s] 53%|█████▎ | 1243/2338 [04:50<04:10, 4.36it/s] 53%|█████▎ | 1244/2338 [04:50<04:38, 3.92it/s] 53%|█████▎ | 1245/2338 [04:50<04:29, 4.06it/s] 53%|█████▎ | 1246/2338 [04:51<04:13, 4.31it/s] 53%|█████▎ | 1247/2338 [04:51<04:13, 4.31it/s] 53%|█████▎ | 1248/2338 [04:51<03:57, 4.58it/s] 53%|█████▎ | 1249/2338 [04:51<03:39, 4.97it/s] 53%|█████▎ | 1250/2338 [04:51<03:46, 4.80it/s] 54%|█████▎ | 1251/2338 [04:52<03:51, 4.70it/s] 54%|█████▎ | 1252/2338 [04:52<03:51, 4.68it/s] 54%|█████▎ | 1253/2338 [04:52<04:01, 4.50it/s] 54%|█████▎ | 1254/2338 [04:52<03:56, 4.59it/s] 54%|█████▎ | 1255/2338 [04:53<03:52, 4.66it/s] 54%|█████▎ | 1256/2338 [04:53<04:01, 4.47it/s] 54%|█████▍ | 1257/2338 [04:53<03:52, 4.65it/s] 54%|█████▍ | 1258/2338 [04:53<04:02, 4.45it/s] 54%|█████▍ | 1259/2338 [04:53<04:06, 4.37it/s] 54%|█████▍ | 1260/2338 [04:54<04:00, 4.49it/s] 54%|█████▍ | 1261/2338 [04:54<04:11, 4.29it/s] 54%|█████▍ | 1262/2338 [04:54<03:52, 4.63it/s] 54%|█████▍ | 1263/2338 [04:54<04:04, 4.40it/s] 54%|█████▍ | 1264/2338 [04:55<04:03, 4.41it/s] 54%|█████▍ | 1265/2338 [04:55<03:57, 4.52it/s] 54%|█████▍ | 1266/2338 [04:55<04:04, 4.39it/s] 54%|█████▍ | 1267/2338 [04:55<04:07, 4.33it/s] 54%|█████▍ | 1268/2338 [04:56<04:20, 4.11it/s] 54%|█████▍ | 1269/2338 [04:56<04:08, 4.31it/s] 54%|█████▍ | 1270/2338 [04:56<04:15, 4.17it/s] 54%|█████▍ | 1271/2338 [04:56<04:13, 4.20it/s] 54%|█████▍ | 1272/2338 [04:57<04:19, 4.11it/s] 54%|█████▍ | 1273/2338 [04:57<04:14, 4.18it/s] 54%|█████▍ | 1274/2338 [04:57<04:21, 4.07it/s] 55%|█████▍ | 1275/2338 [04:57<04:19, 4.09it/s] 55%|█████▍ | 1276/2338 [04:57<04:12, 4.20it/s] 55%|█████▍ | 1277/2338 [04:58<04:09, 4.25it/s] 55%|█████▍ | 1278/2338 [04:58<03:54, 4.52it/s] 55%|█████▍ | 1279/2338 [04:58<04:04, 4.34it/s] 55%|█████▍ | 1280/2338 [04:58<04:19, 4.07it/s] 55%|█████▍ | 1281/2338 [04:59<04:12, 4.18it/s] 55%|█████▍ | 1282/2338 [04:59<04:03, 4.33it/s] 55%|█████▍ | 1283/2338 [04:59<04:09, 4.23it/s] 55%|█████▍ | 1284/2338 [04:59<04:06, 4.28it/s] 55%|█████▍ | 1285/2338 [05:00<04:07, 4.25it/s] 55%|█████▌ | 1286/2338 [05:00<03:58, 4.42it/s] 55%|█████▌ | 1287/2338 [05:00<03:56, 4.45it/s] 55%|█████▌ | 1288/2338 [05:00<03:52, 4.51it/s] 55%|█████▌ | 1289/2338 [05:00<04:06, 4.25it/s] 55%|█████▌ | 1290/2338 [05:01<04:04, 4.28it/s] 55%|█████▌ | 1291/2338 [05:01<04:03, 4.30it/s] 55%|█████▌ | 1292/2338 [05:01<03:37, 4.82it/s] 55%|█████▌ | 1293/2338 [05:01<03:48, 4.58it/s] 55%|█████▌ | 1294/2338 [05:02<03:55, 4.44it/s] 55%|█████▌ | 1295/2338 [05:02<03:58, 4.38it/s] 55%|█████▌ | 1296/2338 [05:02<03:57, 4.39it/s] 55%|█████▌ | 1297/2338 [05:02<03:57, 4.39it/s] 56%|█████▌ | 1298/2338 [05:02<03:53, 4.45it/s] 56%|█████▌ | 1299/2338 [05:03<03:56, 4.39it/s] 56%|█████▌ | 1300/2338 [05:03<04:00, 4.31it/s] 56%|█████▌ | 1301/2338 [05:03<03:57, 4.38it/s] 56%|█████▌ | 1302/2338 [05:03<03:55, 4.39it/s] 56%|█████▌ | 1303/2338 [05:04<03:45, 4.59it/s] 56%|█████▌ | 1304/2338 [05:04<04:00, 4.30it/s] 56%|█████▌ | 1305/2338 [05:04<04:13, 4.08it/s] 56%|█████▌ | 1306/2338 [05:04<04:05, 4.20it/s] 56%|█████▌ | 1307/2338 [05:05<03:59, 4.30it/s] 56%|█████▌ | 1308/2338 [05:05<03:47, 4.54it/s] 56%|█████▌ | 1309/2338 [05:05<03:33, 4.83it/s] 56%|█████▌ | 1310/2338 [05:05<03:48, 4.50it/s] 56%|█████▌ | 1311/2338 [05:05<04:05, 4.18it/s] 56%|█████▌ | 1312/2338 [05:06<03:55, 4.35it/s] 56%|█████▌ | 1313/2338 [05:06<04:14, 4.03it/s] 56%|█████▌ | 1314/2338 [05:06<04:16, 3.98it/s] 56%|█████▌ | 1315/2338 [05:06<04:12, 4.05it/s] 56%|█████▋ | 1316/2338 [05:07<04:10, 4.07it/s] 56%|█████▋ | 1317/2338 [05:07<04:05, 4.15it/s] 56%|█████▋ | 1318/2338 [05:07<04:03, 4.19it/s] 56%|█████▋ | 1319/2338 [05:07<04:02, 4.21it/s] 56%|█████▋ | 1320/2338 [05:08<03:49, 4.44it/s] 57%|█████▋ | 1321/2338 [05:08<04:12, 4.03it/s] 57%|█████▋ | 1322/2338 [05:08<04:36, 3.67it/s] 57%|█████▋ | 1323/2338 [05:08<04:14, 3.98it/s] 57%|█████▋ | 1324/2338 [05:09<03:58, 4.26it/s] 57%|█████▋ | 1325/2338 [05:09<03:54, 4.32it/s] 57%|█████▋ | 1326/2338 [05:09<03:52, 4.35it/s] 57%|█████▋ | 1327/2338 [05:09<04:02, 4.17it/s] 57%|█████▋ | 1328/2338 [05:10<04:00, 4.20it/s] 57%|█████▋ | 1329/2338 [05:10<04:05, 4.10it/s] 57%|█████▋ | 1330/2338 [05:10<04:14, 3.96it/s] 57%|█████▋ | 1331/2338 [05:10<04:10, 4.02it/s] 57%|█████▋ | 1332/2338 [05:11<03:59, 4.20it/s] 57%|█████▋ | 1333/2338 [05:11<03:58, 4.22it/s] 57%|█████▋ | 1334/2338 [05:11<03:54, 4.28it/s] 57%|█████▋ | 1335/2338 [05:11<03:41, 4.54it/s] 57%|█████▋ | 1336/2338 [05:11<03:50, 4.34it/s] 57%|█████▋ | 1337/2338 [05:12<03:40, 4.53it/s] 57%|█████▋ | 1338/2338 [05:12<03:38, 4.57it/s] 57%|█████▋ | 1339/2338 [05:12<03:37, 4.59it/s] 57%|█████▋ | 1340/2338 [05:12<03:43, 4.46it/s] 57%|█████▋ | 1341/2338 [05:13<03:41, 4.49it/s] 57%|█████▋ | 1342/2338 [05:13<04:02, 4.10it/s] 57%|█████▋ | 1343/2338 [05:13<03:59, 4.15it/s] 57%|█████▋ | 1344/2338 [05:13<03:58, 4.17it/s] 58%|█████▊ | 1345/2338 [05:14<03:56, 4.20it/s] 58%|█████▊ | 1346/2338 [05:14<03:53, 4.24it/s] 58%|█████▊ | 1347/2338 [05:14<03:50, 4.31it/s] 58%|█████▊ | 1348/2338 [05:14<04:01, 4.10it/s] 58%|█████▊ | 1349/2338 [05:15<04:03, 4.05it/s] 58%|█████▊ | 1350/2338 [05:15<04:18, 3.83it/s] 58%|█████▊ | 1351/2338 [05:15<04:08, 3.97it/s] 58%|█████▊ | 1352/2338 [05:15<04:09, 3.96it/s] 58%|█████▊ | 1353/2338 [05:16<04:12, 3.91it/s] 58%|█████▊ | 1354/2338 [05:16<03:54, 4.19it/s] 58%|█████▊ | 1355/2338 [05:16<03:51, 4.25it/s] 58%|█████▊ | 1356/2338 [05:16<04:03, 4.03it/s] 58%|█████▊ | 1357/2338 [05:17<03:58, 4.11it/s] 58%|█████▊ | 1358/2338 [05:17<03:57, 4.13it/s] 58%|█████▊ | 1359/2338 [05:17<03:53, 4.19it/s] 58%|█████▊ | 1360/2338 [05:17<03:44, 4.36it/s] 58%|█████▊ | 1361/2338 [05:17<03:52, 4.20it/s] 58%|█████▊ | 1362/2338 [05:18<03:48, 4.28it/s] 58%|█████▊ | 1363/2338 [05:18<03:30, 4.63it/s] 58%|█████▊ | 1364/2338 [05:18<03:40, 4.41it/s] 58%|█████▊ | 1365/2338 [05:18<03:36, 4.50it/s] 58%|█████▊ | 1366/2338 [05:19<03:58, 4.07it/s] 58%|█████▊ | 1367/2338 [05:19<03:54, 4.14it/s] 59%|█████▊ | 1368/2338 [05:19<04:00, 4.03it/s] 59%|█████▊ | 1369/2338 [05:19<03:58, 4.06it/s] 59%|█████▊ | 1370/2338 [05:20<03:47, 4.25it/s] 59%|█████▊ | 1371/2338 [05:20<03:45, 4.29it/s] 59%|█████▊ | 1372/2338 [05:20<03:51, 4.16it/s] 59%|█████▊ | 1373/2338 [05:20<03:19, 4.85it/s] 59%|█████▉ | 1374/2338 [05:20<03:09, 5.09it/s] 59%|█████▉ | 1375/2338 [05:21<03:15, 4.92it/s] 59%|█████▉ | 1376/2338 [05:21<03:30, 4.57it/s] 59%|█████▉ | 1377/2338 [05:21<03:31, 4.54it/s] 59%|█████▉ | 1378/2338 [05:21<03:36, 4.43it/s] 59%|█████▉ | 1379/2338 [05:22<03:38, 4.39it/s] 59%|█████▉ | 1380/2338 [05:22<03:40, 4.35it/s] 59%|█████▉ | 1381/2338 [05:22<03:57, 4.03it/s] 59%|█████▉ | 1382/2338 [05:22<04:00, 3.98it/s] 59%|█████▉ | 1383/2338 [05:23<04:01, 3.95it/s] 59%|█████▉ | 1384/2338 [05:23<03:53, 4.08it/s] 59%|█████▉ | 1385/2338 [05:23<03:38, 4.37it/s] 59%|█████▉ | 1386/2338 [05:23<03:36, 4.39it/s] 59%|█████▉ | 1387/2338 [05:23<03:39, 4.33it/s] 59%|█████▉ | 1388/2338 [05:24<04:01, 3.93it/s] 59%|█████▉ | 1389/2338 [05:24<03:34, 4.42it/s] 59%|█████▉ | 1390/2338 [05:24<03:44, 4.23it/s] 59%|█████▉ | 1391/2338 [05:24<03:38, 4.33it/s] 60%|█████▉ | 1392/2338 [05:25<03:24, 4.63it/s] 60%|█████▉ | 1393/2338 [05:25<03:10, 4.96it/s] 60%|█████▉ | 1394/2338 [05:25<03:27, 4.56it/s] 60%|█████▉ | 1395/2338 [05:25<03:25, 4.59it/s] 60%|█████▉ | 1396/2338 [05:25<03:23, 4.64it/s] 60%|█████▉ | 1397/2338 [05:26<03:33, 4.41it/s] 60%|█████▉ | 1398/2338 [05:26<03:31, 4.45it/s] 60%|█████▉ | 1399/2338 [05:26<03:27, 4.52it/s] 60%|█████▉ | 1400/2338 [05:26<03:38, 4.29it/s] 60%|█████▉ | 1401/2338 [05:27<03:34, 4.37it/s] 60%|█████▉ | 1402/2338 [05:27<03:32, 4.41it/s] 60%|██████ | 1403/2338 [05:27<03:27, 4.52it/s] 60%|██████ | 1404/2338 [05:27<03:31, 4.43it/s] 60%|██████ | 1405/2338 [05:27<03:31, 4.42it/s] 60%|██████ | 1406/2338 [05:28<03:36, 4.30it/s] 60%|██████ | 1407/2338 [05:28<03:38, 4.26it/s] 60%|██████ | 1408/2338 [05:28<03:40, 4.21it/s] 60%|██████ | 1409/2338 [05:28<03:31, 4.38it/s] 60%|██████ | 1410/2338 [05:29<03:31, 4.39it/s] 60%|██████ | 1411/2338 [05:29<03:43, 4.15it/s] 60%|██████ | 1412/2338 [05:29<03:40, 4.21it/s] 60%|██████ | 1413/2338 [05:29<03:16, 4.71it/s] 60%|██████ | 1414/2338 [05:30<03:19, 4.63it/s] 61%|██████ | 1415/2338 [05:30<03:20, 4.61it/s] 61%|██████ | 1416/2338 [05:30<03:22, 4.56it/s] 61%|██████ | 1417/2338 [05:30<03:32, 4.33it/s] 61%|██████ | 1418/2338 [05:30<03:37, 4.22it/s] 61%|██████ | 1419/2338 [05:31<03:36, 4.25it/s] 61%|██████ | 1420/2338 [05:31<03:32, 4.32it/s] 61%|██████ | 1421/2338 [05:31<03:38, 4.19it/s] 61%|██████ | 1422/2338 [05:31<03:32, 4.31it/s] 61%|██████ | 1423/2338 [05:32<03:23, 4.51it/s] 61%|██████ | 1424/2338 [05:32<03:22, 4.51it/s] 61%|██████ | 1425/2338 [05:32<03:26, 4.41it/s] 61%|██████ | 1426/2338 [05:32<03:43, 4.08it/s] 61%|██████ | 1427/2338 [05:33<03:49, 3.96it/s] 61%|██████ | 1428/2338 [05:33<04:00, 3.78it/s] 61%|██████ | 1429/2338 [05:33<03:50, 3.94it/s] 61%|██████ | 1430/2338 [05:33<03:45, 4.03it/s] 61%|██████ | 1431/2338 [05:34<03:17, 4.59it/s] 61%|██████ | 1432/2338 [05:34<03:15, 4.63it/s] 61%|██████▏ | 1433/2338 [05:34<03:22, 4.46it/s] 61%|██████▏ | 1434/2338 [05:34<03:22, 4.45it/s] 61%|██████▏ | 1435/2338 [05:34<03:19, 4.54it/s] 61%|██████▏ | 1436/2338 [05:35<03:18, 4.54it/s] 61%|██████▏ | 1437/2338 [05:35<03:18, 4.55it/s] 62%|██████▏ | 1438/2338 [05:35<03:28, 4.31it/s] 62%|██████▏ | 1439/2338 [05:35<03:46, 3.97it/s] 62%|██████▏ | 1440/2338 [05:36<03:41, 4.06it/s] 62%|██████▏ | 1441/2338 [05:36<03:39, 4.08it/s] 62%|██████▏ | 1442/2338 [05:36<03:31, 4.23it/s] 62%|██████▏ | 1443/2338 [05:36<03:02, 4.90it/s] 62%|██████▏ | 1444/2338 [05:37<03:15, 4.56it/s] 62%|██████▏ | 1445/2338 [05:37<03:23, 4.38it/s] 62%|██████▏ | 1446/2338 [05:37<03:41, 4.03it/s] 62%|██████▏ | 1447/2338 [05:37<03:44, 3.97it/s] 62%|██████▏ | 1448/2338 [05:38<03:30, 4.23it/s] 62%|██████▏ | 1449/2338 [05:38<03:16, 4.52it/s] 62%|██████▏ | 1450/2338 [05:38<03:21, 4.41it/s] 62%|██████▏ | 1451/2338 [05:38<03:32, 4.18it/s] 62%|██████▏ | 1452/2338 [05:38<03:31, 4.18it/s] 62%|██████▏ | 1453/2338 [05:39<03:26, 4.29it/s] 62%|██████▏ | 1454/2338 [05:39<03:25, 4.31it/s] 62%|██████▏ | 1455/2338 [05:39<03:27, 4.26it/s] 62%|██████▏ | 1456/2338 [05:39<03:34, 4.10it/s] 62%|██████▏ | 1457/2338 [05:40<03:37, 4.05it/s] 62%|██████▏ | 1458/2338 [05:40<03:42, 3.95it/s] 62%|██████▏ | 1459/2338 [05:40<03:47, 3.87it/s] 62%|██████▏ | 1460/2338 [05:40<03:45, 3.89it/s] 62%|██████▏ | 1461/2338 [05:41<03:36, 4.05it/s] 63%|██████▎ | 1462/2338 [05:41<03:40, 3.98it/s] 63%|██████▎ | 1463/2338 [05:41<03:43, 3.92it/s] 63%|██████▎ | 1464/2338 [05:41<03:38, 4.01it/s] 63%|██████▎ | 1465/2338 [05:42<03:32, 4.11it/s] 63%|██████▎ | 1466/2338 [05:42<03:24, 4.27it/s] 63%|██████▎ | 1467/2338 [05:42<03:26, 4.21it/s] 63%|██████▎ | 1468/2338 [05:42<03:26, 4.21it/s] 63%|██████▎ | 1469/2338 [05:43<03:22, 4.29it/s] 63%|██████▎ | 1470/2338 [05:43<03:18, 4.36it/s] 63%|██████▎ | 1471/2338 [05:43<03:26, 4.21it/s] 63%|██████▎ | 1472/2338 [05:43<03:30, 4.12it/s] 63%|██████▎ | 1473/2338 [05:44<03:37, 3.98it/s] 63%|██████▎ | 1474/2338 [05:44<03:38, 3.95it/s] 63%|██████▎ | 1475/2338 [05:44<03:33, 4.04it/s] 63%|██████▎ | 1476/2338 [05:44<03:24, 4.21it/s] 63%|██████▎ | 1477/2338 [05:45<03:22, 4.25it/s] 63%|██████▎ | 1478/2338 [05:45<03:22, 4.24it/s] 63%|██████▎ | 1479/2338 [05:45<03:38, 3.93it/s] 63%|██████▎ | 1480/2338 [05:45<03:27, 4.13it/s] 63%|██████▎ | 1481/2338 [05:45<03:16, 4.36it/s] 63%|██████▎ | 1482/2338 [05:46<03:17, 4.33it/s] 63%|██████▎ | 1483/2338 [05:46<03:23, 4.20it/s] 63%|██████▎ | 1484/2338 [05:46<03:22, 4.23it/s] 64%|██████▎ | 1485/2338 [05:46<03:18, 4.29it/s] 64%|██████▎ | 1486/2338 [05:47<03:16, 4.34it/s] 64%|██████▎ | 1487/2338 [05:47<03:11, 4.43it/s] 64%|██████▎ | 1488/2338 [05:47<03:18, 4.28it/s] 64%|██████▎ | 1489/2338 [05:47<03:09, 4.47it/s] 64%|██████▎ | 1490/2338 [05:48<03:12, 4.41it/s] 64%|██████▍ | 1491/2338 [05:48<03:15, 4.33it/s] 64%|██████▍ | 1492/2338 [05:48<03:19, 4.25it/s] 64%|██████▍ | 1493/2338 [05:48<03:10, 4.44it/s] 64%|██████▍ | 1494/2338 [05:48<03:13, 4.36it/s] 64%|██████▍ | 1495/2338 [05:49<03:10, 4.43it/s] 64%|██████▍ | 1496/2338 [05:49<03:13, 4.35it/s] 64%|██████▍ | 1497/2338 [05:49<02:51, 4.90it/s] 64%|██████▍ | 1498/2338 [05:49<03:03, 4.59it/s] 64%|██████▍ | 1499/2338 [05:50<03:05, 4.53it/s] 64%|██████▍ | 1500/2338 [05:50<03:10, 4.40it/s] 64%|██████▍ | 1501/2338 [05:50<03:17, 4.25it/s] 64%|██████▍ | 1502/2338 [05:50<03:14, 4.30it/s] 64%|██████▍ | 1503/2338 [05:50<03:03, 4.54it/s] 64%|██████▍ | 1504/2338 [05:51<03:14, 4.30it/s] 64%|██████▍ | 1505/2338 [05:51<03:18, 4.19it/s] 64%|██████▍ | 1506/2338 [05:51<03:33, 3.89it/s] 64%|██████▍ | 1507/2338 [05:52<03:29, 3.97it/s] 64%|██████▍ | 1508/2338 [05:52<03:34, 3.88it/s] 65%|██████▍ | 1509/2338 [05:52<03:15, 4.23it/s] 65%|██████▍ | 1510/2338 [05:52<03:21, 4.12it/s] 65%|██████▍ | 1511/2338 [05:52<03:09, 4.37it/s] 65%|██████▍ | 1512/2338 [05:53<03:00, 4.58it/s] 65%|██████▍ | 1513/2338 [05:53<03:04, 4.47it/s] 65%|██████▍ | 1514/2338 [05:53<03:14, 4.25it/s] 65%|██████▍ | 1515/2338 [05:53<03:13, 4.24it/s] 65%|██████▍ | 1516/2338 [05:54<03:15, 4.20it/s] 65%|██████▍ | 1517/2338 [05:54<03:15, 4.21it/s] 65%|██████▍ | 1518/2338 [05:54<03:18, 4.13it/s] 65%|██████▍ | 1519/2338 [05:54<03:20, 4.09it/s] 65%|██████▌ | 1520/2338 [05:55<03:25, 3.98it/s] 65%|██████▌ | 1521/2338 [05:55<03:20, 4.06it/s] 65%|██████▌ | 1522/2338 [05:55<03:22, 4.03it/s] 65%|██████▌ | 1523/2338 [05:55<03:15, 4.17it/s] 65%|██████▌ | 1524/2338 [05:56<03:12, 4.22it/s] 65%|██████▌ | 1525/2338 [05:56<03:08, 4.32it/s] 65%|██████▌ | 1526/2338 [05:56<03:20, 4.04it/s] 65%|██████▌ | 1527/2338 [05:56<03:33, 3.80it/s] 65%|██████▌ | 1528/2338 [05:57<03:44, 3.61it/s] 65%|██████▌ | 1529/2338 [05:57<03:37, 3.72it/s] 65%|██████▌ | 1530/2338 [05:57<03:33, 3.79it/s] 65%|██████▌ | 1531/2338 [05:57<03:30, 3.83it/s] 66%|██████▌ | 1532/2338 [05:58<03:21, 3.99it/s] 66%|██████▌ | 1533/2338 [05:58<03:24, 3.95it/s] 66%|██████▌ | 1534/2338 [05:58<03:27, 3.87it/s] 66%|██████▌ | 1535/2338 [05:58<03:23, 3.94it/s] 66%|██████▌ | 1536/2338 [05:59<03:16, 4.07it/s] 66%|██████▌ | 1537/2338 [05:59<03:12, 4.17it/s] 66%|██████▌ | 1538/2338 [05:59<03:08, 4.25it/s] 66%|██████▌ | 1539/2338 [05:59<03:07, 4.26it/s] 66%|██████▌ | 1540/2338 [06:00<03:04, 4.31it/s] 66%|██████▌ | 1541/2338 [06:00<02:47, 4.77it/s] 66%|██████▌ | 1542/2338 [06:00<02:51, 4.63it/s] 66%|██████▌ | 1543/2338 [06:00<02:54, 4.55it/s] 66%|██████▌ | 1544/2338 [06:00<03:00, 4.40it/s] 66%|██████▌ | 1545/2338 [06:01<02:58, 4.44it/s] 66%|██████▌ | 1546/2338 [06:01<03:09, 4.19it/s] 66%|██████▌ | 1547/2338 [06:01<03:04, 4.28it/s] 66%|██████▌ | 1548/2338 [06:01<03:00, 4.38it/s] 66%|██████▋ | 1549/2338 [06:02<03:01, 4.34it/s] 66%|██████▋ | 1550/2338 [06:02<02:47, 4.71it/s] 66%|██████▋ | 1551/2338 [06:02<03:02, 4.31it/s] 66%|██████▋ | 1552/2338 [06:02<03:09, 4.16it/s] 66%|██████▋ | 1553/2338 [06:03<03:13, 4.05it/s] 66%|██████▋ | 1554/2338 [06:03<03:10, 4.11it/s] 67%|██████▋ | 1555/2338 [06:03<03:11, 4.08it/s] 67%|██████▋ | 1556/2338 [06:03<03:05, 4.23it/s] 67%|██████▋ | 1557/2338 [06:04<03:11, 4.09it/s] 67%|██████▋ | 1558/2338 [06:04<03:16, 3.96it/s] 67%|██████▋ | 1559/2338 [06:04<03:18, 3.93it/s] 67%|██████▋ | 1560/2338 [06:04<03:09, 4.11it/s] 67%|██████▋ | 1561/2338 [06:04<03:05, 4.18it/s] 67%|██████▋ | 1562/2338 [06:05<03:15, 3.96it/s] 67%|██████▋ | 1563/2338 [06:05<03:15, 3.96it/s] 67%|██████▋ | 1564/2338 [06:05<03:10, 4.06it/s] 67%|██████▋ | 1565/2338 [06:06<03:23, 3.80it/s] 67%|██████▋ | 1566/2338 [06:06<03:14, 3.96it/s] 67%|██████▋ | 1567/2338 [06:06<03:13, 3.99it/s] 67%|██████▋ | 1568/2338 [06:06<03:07, 4.10it/s] 67%|██████▋ | 1569/2338 [06:07<03:08, 4.07it/s] 67%|██████▋ | 1570/2338 [06:07<03:02, 4.22it/s] 67%|██████▋ | 1571/2338 [06:07<03:02, 4.19it/s] 67%|██████▋ | 1572/2338 [06:07<02:59, 4.26it/s] 67%|██████▋ | 1573/2338 [06:07<03:05, 4.12it/s] 67%|██████▋ | 1574/2338 [06:08<03:10, 4.01it/s] 67%|██████▋ | 1575/2338 [06:08<02:51, 4.44it/s] 67%|██████▋ | 1576/2338 [06:08<02:56, 4.31it/s] 67%|██████▋ | 1577/2338 [06:08<02:53, 4.40it/s] 67%|██████▋ | 1578/2338 [06:09<02:46, 4.56it/s] 68%|██████▊ | 1579/2338 [06:09<02:36, 4.84it/s] 68%|██████▊ | 1580/2338 [06:09<02:39, 4.74it/s] 68%|██████▊ | 1581/2338 [06:09<02:42, 4.66it/s] 68%|██████▊ | 1582/2338 [06:09<02:45, 4.56it/s] 68%|██████▊ | 1583/2338 [06:10<02:59, 4.21it/s] 68%|██████▊ | 1584/2338 [06:10<02:53, 4.35it/s] 68%|██████▊ | 1585/2338 [06:10<02:58, 4.22it/s] 68%|██████▊ | 1586/2338 [06:10<02:50, 4.41it/s] 68%|██████▊ | 1587/2338 [06:11<02:45, 4.54it/s] 68%|██████▊ | 1588/2338 [06:11<02:41, 4.65it/s] 68%|██████▊ | 1589/2338 [06:11<02:47, 4.46it/s] 68%|██████▊ | 1590/2338 [06:11<02:50, 4.39it/s] 68%|██████▊ | 1591/2338 [06:11<02:52, 4.32it/s] 68%|██████▊ | 1592/2338 [06:12<02:44, 4.53it/s] 68%|██████▊ | 1593/2338 [06:12<02:47, 4.45it/s] 68%|██████▊ | 1594/2338 [06:12<02:51, 4.35it/s] 68%|██████▊ | 1595/2338 [06:12<02:51, 4.34it/s] 68%|██████▊ | 1596/2338 [06:13<02:52, 4.29it/s] 68%|██████▊ | 1597/2338 [06:13<02:48, 4.41it/s] 68%|██████▊ | 1598/2338 [06:13<02:47, 4.43it/s] 68%|██████▊ | 1599/2338 [06:13<02:45, 4.48it/s] 68%|██████▊ | 1600/2338 [06:14<02:48, 4.37it/s] 68%|██████▊ | 1601/2338 [06:14<02:33, 4.79it/s] 69%|██████▊ | 1602/2338 [06:14<02:41, 4.54it/s] 69%|██████▊ | 1603/2338 [06:14<02:46, 4.41it/s] 69%|██████▊ | 1604/2338 [06:14<03:03, 4.00it/s] 69%|██████▊ | 1605/2338 [06:15<03:05, 3.96it/s] 69%|██████▊ | 1606/2338 [06:15<02:59, 4.08it/s] 69%|██████▊ | 1607/2338 [06:15<02:51, 4.25it/s] 69%|██████▉ | 1608/2338 [06:15<02:50, 4.28it/s] 69%|██████▉ | 1609/2338 [06:16<02:47, 4.35it/s] 69%|██████▉ | 1610/2338 [06:16<02:56, 4.11it/s] 69%|██████▉ | 1611/2338 [06:16<02:55, 4.14it/s] 69%|██████▉ | 1612/2338 [06:16<02:54, 4.16it/s] 69%|██████▉ | 1613/2338 [06:17<02:49, 4.28it/s] 69%|██████▉ | 1614/2338 [06:17<02:51, 4.22it/s] 69%|██████▉ | 1615/2338 [06:17<02:46, 4.35it/s] 69%|██████▉ | 1616/2338 [06:17<02:47, 4.30it/s] 69%|██████▉ | 1617/2338 [06:18<02:47, 4.31it/s] 69%|██████▉ | 1618/2338 [06:18<02:44, 4.39it/s] 69%|██████▉ | 1619/2338 [06:18<02:47, 4.29it/s] 69%|██████▉ | 1620/2338 [06:18<02:42, 4.41it/s] 69%|██████▉ | 1621/2338 [06:18<02:44, 4.36it/s] 69%|██████▉ | 1622/2338 [06:19<02:42, 4.41it/s] 69%|██████▉ | 1623/2338 [06:19<02:43, 4.37it/s] 69%|██████▉ | 1624/2338 [06:19<02:42, 4.40it/s] 70%|██████▉ | 1625/2338 [06:19<02:40, 4.45it/s] 70%|██████▉ | 1626/2338 [06:20<02:38, 4.48it/s] 70%|██████▉ | 1627/2338 [06:20<02:41, 4.40it/s] 70%|██████▉ | 1628/2338 [06:20<02:45, 4.29it/s] 70%|██████▉ | 1629/2338 [06:20<02:41, 4.40it/s] 70%|██████▉ | 1630/2338 [06:20<02:31, 4.66it/s] 70%|██████▉ | 1631/2338 [06:21<02:23, 4.94it/s] 70%|██████▉ | 1632/2338 [06:21<02:25, 4.86it/s] 70%|██████▉ | 1633/2338 [06:21<02:28, 4.76it/s] 70%|██████▉ | 1634/2338 [06:21<02:29, 4.71it/s] 70%|██████▉ | 1635/2338 [06:21<02:28, 4.74it/s] 70%|██████▉ | 1636/2338 [06:22<02:53, 4.04it/s] 70%|███████ | 1637/2338 [06:22<02:54, 4.02it/s] 70%|███████ | 1638/2338 [06:22<02:53, 4.04it/s] 70%|███████ | 1639/2338 [06:23<02:54, 4.01it/s] 70%|███████ | 1640/2338 [06:23<02:52, 4.04it/s] 70%|███████ | 1641/2338 [06:23<02:47, 4.15it/s] 70%|███████ | 1642/2338 [06:23<02:53, 4.00it/s] 70%|███████ | 1643/2338 [06:23<02:44, 4.22it/s] 70%|███████ | 1644/2338 [06:24<02:43, 4.23it/s] 70%|███████ | 1645/2338 [06:24<02:42, 4.27it/s] 70%|███████ | 1646/2338 [06:24<02:43, 4.22it/s] 70%|███████ | 1647/2338 [06:24<02:32, 4.54it/s] 70%|███████ | 1648/2338 [06:25<02:31, 4.56it/s] 71%|███████ | 1649/2338 [06:25<02:32, 4.52it/s] 71%|███████ | 1650/2338 [06:25<02:39, 4.33it/s] 71%|███████ | 1651/2338 [06:25<02:49, 4.05it/s] 71%|███████ | 1652/2338 [06:26<02:48, 4.08it/s] 71%|███████ | 1653/2338 [06:26<02:44, 4.16it/s] 71%|███████ | 1654/2338 [06:26<02:37, 4.35it/s] 71%|███████ | 1655/2338 [06:26<02:28, 4.60it/s] 71%|███████ | 1656/2338 [06:26<02:17, 4.97it/s] 71%|███████ | 1657/2338 [06:27<02:24, 4.72it/s] 71%|███████ | 1658/2338 [06:27<02:31, 4.48it/s] 71%|███████ | 1659/2338 [06:27<02:26, 4.65it/s] 71%|███████ | 1660/2338 [06:27<02:30, 4.51it/s] 71%|███████ | 1661/2338 [06:28<02:38, 4.28it/s] 71%|███████ | 1662/2338 [06:28<02:49, 3.98it/s] 71%|███████ | 1663/2338 [06:28<02:52, 3.91it/s] 71%|███████ | 1664/2338 [06:28<02:45, 4.07it/s] 71%|███████ | 1665/2338 [06:29<02:45, 4.06it/s] 71%|███████▏ | 1666/2338 [06:29<02:45, 4.06it/s] 71%|███████▏ | 1667/2338 [06:29<02:39, 4.21it/s] 71%|███████▏ | 1668/2338 [06:29<02:35, 4.30it/s] 71%|███████▏ | 1669/2338 [06:29<02:30, 4.44it/s] 71%|███████▏ | 1670/2338 [06:30<02:17, 4.84it/s] 71%|███████▏ | 1671/2338 [06:30<02:36, 4.27it/s] 72%|███████▏ | 1672/2338 [06:30<02:34, 4.31it/s] 72%|███████▏ | 1673/2338 [06:30<02:31, 4.38it/s] 72%|███████▏ | 1674/2338 [06:31<02:29, 4.46it/s] 72%|███████▏ | 1675/2338 [06:31<02:32, 4.36it/s] 72%|███████▏ | 1676/2338 [06:31<02:32, 4.34it/s] 72%|███████▏ | 1677/2338 [06:31<02:33, 4.32it/s] 72%|███████▏ | 1678/2338 [06:32<02:31, 4.34it/s] 72%|███████▏ | 1679/2338 [06:32<02:21, 4.66it/s] 72%|███████▏ | 1680/2338 [06:32<02:27, 4.46it/s] 72%|███████▏ | 1681/2338 [06:32<02:29, 4.38it/s] 72%|███████▏ | 1682/2338 [06:32<02:30, 4.37it/s] 72%|███████▏ | 1683/2338 [06:33<02:28, 4.42it/s] 72%|███████▏ | 1684/2338 [06:33<02:25, 4.50it/s] 72%|███████▏ | 1685/2338 [06:33<02:28, 4.41it/s] 72%|███████▏ | 1686/2338 [06:33<02:27, 4.42it/s] 72%|███████▏ | 1687/2338 [06:34<02:27, 4.42it/s] 72%|███████▏ | 1688/2338 [06:34<02:27, 4.42it/s] 72%|███████▏ | 1689/2338 [06:34<02:29, 4.35it/s] 72%|███████▏ | 1690/2338 [06:34<02:26, 4.43it/s] 72%|███████▏ | 1691/2338 [06:34<02:28, 4.37it/s] 72%|███████▏ | 1692/2338 [06:35<02:29, 4.31it/s] 72%|███████▏ | 1693/2338 [06:35<02:27, 4.37it/s] 72%|███████▏ | 1694/2338 [06:35<02:34, 4.17it/s] 72%|███████▏ | 1695/2338 [06:35<02:31, 4.25it/s] 73%|███████▎ | 1696/2338 [06:36<02:28, 4.33it/s] 73%|███████▎ | 1697/2338 [06:36<02:24, 4.43it/s] 73%|███████▎ | 1698/2338 [06:36<02:24, 4.42it/s] 73%|███████▎ | 1699/2338 [06:36<02:20, 4.55it/s] 73%|███████▎ | 1700/2338 [06:37<02:18, 4.62it/s] 73%|███████▎ | 1701/2338 [06:37<02:24, 4.41it/s] 73%|███████▎ | 1702/2338 [06:37<02:26, 4.34it/s] 73%|███████▎ | 1703/2338 [06:37<02:28, 4.27it/s] 73%|███████▎ | 1704/2338 [06:37<02:27, 4.28it/s] 73%|███████▎ | 1705/2338 [06:38<02:28, 4.27it/s] 73%|███████▎ | 1706/2338 [06:38<02:19, 4.51it/s] 73%|███████▎ | 1707/2338 [06:38<02:17, 4.59it/s] 73%|███████▎ | 1708/2338 [06:38<02:20, 4.47it/s] 73%|███████▎ | 1709/2338 [06:39<02:22, 4.40it/s] 73%|███████▎ | 1710/2338 [06:39<02:19, 4.51it/s] 73%|███████▎ | 1711/2338 [06:39<02:13, 4.71it/s] 73%|███████▎ | 1712/2338 [06:39<02:06, 4.94it/s] 73%|███████▎ | 1713/2338 [06:39<02:10, 4.78it/s] 73%|███████▎ | 1714/2338 [06:40<02:17, 4.52it/s] 73%|███████▎ | 1715/2338 [06:40<02:14, 4.62it/s] 73%|███████▎ | 1716/2338 [06:40<02:22, 4.38it/s] 73%|███████▎ | 1717/2338 [06:40<02:24, 4.30it/s] 73%|███████▎ | 1718/2338 [06:41<02:31, 4.08it/s] 74%|███████▎ | 1719/2338 [06:41<02:20, 4.40it/s] 74%|███████▎ | 1720/2338 [06:41<02:28, 4.15it/s] 74%|███████▎ | 1721/2338 [06:41<02:18, 4.45it/s] 74%|███████▎ | 1722/2338 [06:41<02:14, 4.57it/s] 74%|███████▎ | 1723/2338 [06:42<02:20, 4.37it/s] 74%|███████▎ | 1724/2338 [06:42<02:23, 4.27it/s] 74%|███████▍ | 1725/2338 [06:42<02:19, 4.39it/s] 74%|███████▍ | 1726/2338 [06:42<02:22, 4.31it/s] 74%|███████▍ | 1727/2338 [06:43<02:28, 4.10it/s] 74%|███████▍ | 1728/2338 [06:43<02:29, 4.09it/s] 74%|███████▍ | 1729/2338 [06:43<02:28, 4.09it/s] 74%|███████▍ | 1730/2338 [06:43<02:32, 3.97it/s] 74%|███████▍ | 1731/2338 [06:44<02:43, 3.71it/s] 74%|███████▍ | 1732/2338 [06:44<02:36, 3.87it/s] 74%|███████▍ | 1733/2338 [06:44<02:34, 3.91it/s] 74%|███████▍ | 1734/2338 [06:44<02:30, 4.02it/s] 74%|███████▍ | 1735/2338 [06:45<02:28, 4.06it/s] 74%|███████▍ | 1736/2338 [06:45<02:28, 4.06it/s] 74%|███████▍ | 1737/2338 [06:45<02:22, 4.23it/s] 74%|███████▍ | 1738/2338 [06:45<02:21, 4.23it/s] 74%|███████▍ | 1739/2338 [06:46<02:22, 4.20it/s] 74%|███████▍ | 1740/2338 [06:46<02:18, 4.33it/s] 74%|███████▍ | 1741/2338 [06:46<02:19, 4.29it/s] 75%|███████▍ | 1742/2338 [06:46<02:16, 4.36it/s] 75%|███████▍ | 1743/2338 [06:47<02:18, 4.31it/s] 75%|███████▍ | 1744/2338 [06:47<02:22, 4.18it/s] 75%|███████▍ | 1745/2338 [06:47<02:17, 4.32it/s] 75%|███████▍ | 1746/2338 [06:47<02:23, 4.13it/s] 75%|███████▍ | 1747/2338 [06:48<02:29, 3.95it/s] 75%|███████▍ | 1748/2338 [06:48<02:21, 4.17it/s] 75%|███████▍ | 1749/2338 [06:48<02:22, 4.14it/s] 75%|███████▍ | 1750/2338 [06:48<02:20, 4.18it/s] 75%|███████▍ | 1751/2338 [06:48<02:17, 4.26it/s] 75%|███████▍ | 1752/2338 [06:49<02:25, 4.03it/s] 75%|███████▍ | 1753/2338 [06:49<02:22, 4.09it/s] 75%|███████▌ | 1754/2338 [06:49<02:24, 4.05it/s] 75%|███████▌ | 1755/2338 [06:49<02:23, 4.05it/s] 75%|███████▌ | 1756/2338 [06:50<02:22, 4.07it/s] 75%|███████▌ | 1757/2338 [06:50<02:28, 3.91it/s] 75%|███████▌ | 1758/2338 [06:50<02:26, 3.95it/s] 75%|███████▌ | 1759/2338 [06:51<02:27, 3.93it/s] 75%|███████▌ | 1760/2338 [06:51<02:15, 4.27it/s] 75%|███████▌ | 1761/2338 [06:51<02:14, 4.28it/s] 75%|███████▌ | 1762/2338 [06:51<02:21, 4.06it/s] 75%|███████▌ | 1763/2338 [06:51<02:19, 4.13it/s] 75%|███████▌ | 1764/2338 [06:52<02:12, 4.33it/s] 75%|███████▌ | 1765/2338 [06:52<02:13, 4.29it/s] 76%|███████▌ | 1766/2338 [06:52<02:12, 4.31it/s] 76%|███████▌ | 1767/2338 [06:52<02:10, 4.37it/s] 76%|███████▌ | 1768/2338 [06:53<02:08, 4.42it/s] 76%|███████▌ | 1769/2338 [06:53<02:05, 4.53it/s] 76%|███████▌ | 1770/2338 [06:53<02:07, 4.45it/s] 76%|███████▌ | 1771/2338 [06:53<02:06, 4.47it/s] 76%|███████▌ | 1772/2338 [06:54<02:17, 4.12it/s] 76%|███████▌ | 1773/2338 [06:54<02:22, 3.97it/s] 76%|███████▌ | 1774/2338 [06:54<02:19, 4.05it/s] 76%|███████▌ | 1775/2338 [06:54<02:17, 4.11it/s] 76%|███████▌ | 1776/2338 [06:55<02:16, 4.12it/s] 76%|███████▌ | 1777/2338 [06:55<02:12, 4.22it/s] 76%|███████▌ | 1778/2338 [06:55<02:12, 4.23it/s] 76%|███████▌ | 1779/2338 [06:55<02:07, 4.37it/s] 76%|███████▌ | 1780/2338 [06:55<02:07, 4.38it/s] 76%|███████▌ | 1781/2338 [06:56<02:12, 4.20it/s] 76%|███████▌ | 1782/2338 [06:56<02:05, 4.44it/s] 76%|███████▋ | 1783/2338 [06:56<02:03, 4.48it/s] 76%|███████▋ | 1784/2338 [06:56<02:04, 4.45it/s] 76%|███████▋ | 1785/2338 [06:57<02:05, 4.40it/s] 76%|███████▋ | 1786/2338 [06:57<02:14, 4.10it/s] 76%|███████▋ | 1787/2338 [06:57<02:10, 4.22it/s] 76%|███████▋ | 1788/2338 [06:57<02:08, 4.28it/s] 77%|███████▋ | 1789/2338 [06:57<02:05, 4.36it/s] 77%|███████▋ | 1790/2338 [06:58<02:03, 4.44it/s] 77%|███████▋ | 1791/2338 [06:58<02:16, 4.01it/s] 77%|███████▋ | 1792/2338 [06:58<02:16, 4.00it/s] 77%|███████▋ | 1793/2338 [06:58<02:14, 4.05it/s] 77%|███████▋ | 1794/2338 [06:59<02:16, 3.98it/s] 77%|███████▋ | 1795/2338 [06:59<02:13, 4.07it/s] 77%|███████▋ | 1796/2338 [06:59<02:10, 4.17it/s] 77%|███████▋ | 1797/2338 [06:59<02:10, 4.16it/s] 77%|███████▋ | 1798/2338 [07:00<02:07, 4.24it/s] 77%|███████▋ | 1799/2338 [07:00<02:05, 4.29it/s] 77%|███████▋ | 1800/2338 [07:00<02:01, 4.43it/s] 77%|███████▋ | 1801/2338 [07:00<02:07, 4.22it/s] 77%|███████▋ | 1802/2338 [07:01<02:02, 4.37it/s] 77%|███████▋ | 1803/2338 [07:01<02:09, 4.12it/s] 77%|███████▋ | 1804/2338 [07:01<02:16, 3.90it/s] 77%|███████▋ | 1805/2338 [07:01<02:21, 3.78it/s] 77%|███████▋ | 1806/2338 [07:02<02:16, 3.89it/s] 77%|███████▋ | 1807/2338 [07:02<02:13, 3.98it/s] 77%|███████▋ | 1808/2338 [07:02<02:12, 3.99it/s] 77%|███████▋ | 1809/2338 [07:02<02:14, 3.95it/s] 77%|███████▋ | 1810/2338 [07:03<02:12, 3.99it/s] 77%|███████▋ | 1811/2338 [07:03<02:13, 3.94it/s] 78%|███████▊ | 1812/2338 [07:03<02:10, 4.04it/s] 78%|███████▊ | 1813/2338 [07:03<02:08, 4.09it/s] 78%|███████▊ | 1814/2338 [07:04<02:05, 4.17it/s] 78%|███████▊ | 1815/2338 [07:04<02:07, 4.09it/s] 78%|███████▊ | 1816/2338 [07:04<02:03, 4.21it/s] 78%|███████▊ | 1817/2338 [07:04<02:02, 4.25it/s] 78%|███████▊ | 1818/2338 [07:05<02:01, 4.30it/s] 78%|███████▊ | 1819/2338 [07:05<01:58, 4.39it/s] 78%|███████▊ | 1820/2338 [07:05<02:05, 4.14it/s] 78%|███████▊ | 1821/2338 [07:05<02:07, 4.06it/s] 78%|███████▊ | 1822/2338 [07:06<02:09, 3.98it/s] 78%|███████▊ | 1823/2338 [07:06<02:08, 4.01it/s] 78%|███████▊ | 1824/2338 [07:06<02:04, 4.12it/s] 78%|███████▊ | 1825/2338 [07:06<01:52, 4.55it/s] 78%|███████▊ | 1826/2338 [07:06<01:54, 4.46it/s] 78%|███████▊ | 1827/2338 [07:07<01:54, 4.48it/s] 78%|███████▊ | 1828/2338 [07:07<01:44, 4.87it/s] 78%|███████▊ | 1829/2338 [07:07<01:38, 5.18it/s] 78%|███████▊ | 1830/2338 [07:07<01:56, 4.36it/s] 78%|███████▊ | 1831/2338 [07:07<01:49, 4.64it/s] 78%|███████▊ | 1832/2338 [07:08<01:54, 4.41it/s] 78%|███████▊ | 1833/2338 [07:08<02:00, 4.18it/s] 78%|███████▊ | 1834/2338 [07:08<02:04, 4.05it/s] 78%|███████▊ | 1835/2338 [07:09<02:02, 4.12it/s] 79%|███████▊ | 1836/2338 [07:09<02:03, 4.07it/s] 79%|███████▊ | 1837/2338 [07:09<02:01, 4.13it/s] 79%|███████▊ | 1838/2338 [07:09<02:06, 3.94it/s] 79%|███████▊ | 1839/2338 [07:09<02:00, 4.15it/s] 79%|███████▊ | 1840/2338 [07:10<01:58, 4.20it/s] 79%|███████▊ | 1841/2338 [07:10<02:00, 4.14it/s] 79%|███████▉ | 1842/2338 [07:10<02:04, 4.00it/s] 79%|███████▉ | 1843/2338 [07:10<02:04, 3.98it/s] 79%|███████▉ | 1844/2338 [07:11<02:03, 3.99it/s] 79%|███████▉ | 1845/2338 [07:11<02:00, 4.09it/s] 79%|███████▉ | 1846/2338 [07:11<01:56, 4.21it/s] 79%|███████▉ | 1847/2338 [07:11<01:56, 4.21it/s] 79%|███████▉ | 1848/2338 [07:12<01:53, 4.32it/s] 79%|███████▉ | 1849/2338 [07:12<01:51, 4.39it/s] 79%|███████▉ | 1850/2338 [07:12<01:49, 4.44it/s] 79%|███████▉ | 1851/2338 [07:12<01:53, 4.29it/s] 79%|███████▉ | 1852/2338 [07:13<01:51, 4.34it/s] 79%|███████▉ | 1853/2338 [07:13<01:50, 4.39it/s] 79%|███████▉ | 1854/2338 [07:13<01:51, 4.32it/s] 79%|███████▉ | 1855/2338 [07:13<01:57, 4.11it/s] 79%|███████▉ | 1856/2338 [07:14<02:03, 3.90it/s] 79%|███████▉ | 1857/2338 [07:14<02:01, 3.95it/s] 79%|███████▉ | 1858/2338 [07:14<01:59, 4.02it/s] 80%|███████▉ | 1859/2338 [07:14<01:58, 4.03it/s] 80%|███████▉ | 1860/2338 [07:15<01:52, 4.24it/s] 80%|███████▉ | 1861/2338 [07:15<01:52, 4.23it/s] 80%|███████▉ | 1862/2338 [07:15<01:58, 4.00it/s] 80%|███████▉ | 1863/2338 [07:15<01:59, 3.96it/s] 80%|███████▉ | 1864/2338 [07:16<01:55, 4.10it/s] 80%|███████▉ | 1865/2338 [07:16<01:52, 4.21it/s] 80%|███████▉ | 1866/2338 [07:16<01:52, 4.21it/s] 80%|███████▉ | 1867/2338 [07:16<01:51, 4.21it/s] 80%|███████▉ | 1868/2338 [07:16<01:52, 4.18it/s] 80%|███████▉ | 1869/2338 [07:17<01:54, 4.09it/s] 80%|███████▉ | 1870/2338 [07:17<01:53, 4.11it/s] 80%|████████ | 1871/2338 [07:17<01:52, 4.15it/s] 80%|████████ | 1872/2338 [07:17<01:53, 4.12it/s] 80%|████████ | 1873/2338 [07:18<01:45, 4.43it/s] 80%|████████ | 1874/2338 [07:18<01:45, 4.40it/s] 80%|████████ | 1875/2338 [07:18<01:51, 4.15it/s] 80%|████████ | 1876/2338 [07:18<01:54, 4.05it/s] 80%|████████ | 1877/2338 [07:19<01:53, 4.07it/s] 80%|████████ | 1878/2338 [07:19<01:50, 4.17it/s] 80%|████████ | 1879/2338 [07:19<01:46, 4.30it/s] 80%|████████ | 1880/2338 [07:19<01:46, 4.28it/s] 80%|████████ | 1881/2338 [07:19<01:38, 4.63it/s] 80%|████████ | 1882/2338 [07:20<01:39, 4.58it/s] 81%|████████ | 1883/2338 [07:20<01:43, 4.40it/s] 81%|████████ | 1884/2338 [07:20<01:37, 4.64it/s] 81%|████████ | 1885/2338 [07:20<01:41, 4.47it/s] 81%|████████ | 1886/2338 [07:21<01:48, 4.16it/s] 81%|████████ | 1887/2338 [07:21<01:47, 4.19it/s] 81%|████████ | 1888/2338 [07:21<01:51, 4.05it/s] 81%|████████ | 1889/2338 [07:21<01:53, 3.97it/s] 81%|████████ | 1890/2338 [07:22<01:51, 4.03it/s] 81%|████████ | 1891/2338 [07:22<01:55, 3.86it/s] 81%|████████ | 1892/2338 [07:22<02:01, 3.68it/s] 81%|████████ | 1893/2338 [07:23<01:56, 3.82it/s] 81%|████████ | 1894/2338 [07:23<01:55, 3.83it/s] 81%|████████ | 1895/2338 [07:23<01:55, 3.84it/s] 81%|████████ | 1896/2338 [07:23<01:52, 3.92it/s] 81%|████████ | 1897/2338 [07:24<01:58, 3.73it/s] 81%|████████ | 1898/2338 [07:24<01:53, 3.87it/s] 81%|████████ | 1899/2338 [07:24<01:52, 3.90it/s] 81%|████████▏ | 1900/2338 [07:24<01:48, 4.04it/s] 81%|████████▏ | 1901/2338 [07:25<01:46, 4.09it/s] 81%|████████▏ | 1902/2338 [07:25<01:44, 4.17it/s] 81%|████████▏ | 1903/2338 [07:25<01:40, 4.34it/s] 81%|████████▏ | 1904/2338 [07:25<01:45, 4.13it/s] 81%|████████▏ | 1905/2338 [07:25<01:45, 4.12it/s] 82%|████████▏ | 1906/2338 [07:26<01:49, 3.94it/s] 82%|████████▏ | 1907/2338 [07:26<01:46, 4.06it/s] 82%|████████▏ | 1908/2338 [07:26<01:44, 4.13it/s] 82%|████████▏ | 1909/2338 [07:26<01:49, 3.91it/s] 82%|████████▏ | 1910/2338 [07:27<01:44, 4.10it/s] 82%|████████▏ | 1911/2338 [07:27<01:41, 4.22it/s] 82%|████████▏ | 1912/2338 [07:27<01:37, 4.35it/s] 82%|████████▏ | 1913/2338 [07:27<01:37, 4.35it/s] 82%|████████▏ | 1914/2338 [07:28<01:37, 4.35it/s] 82%|████████▏ | 1915/2338 [07:28<01:27, 4.86it/s] 82%|████████▏ | 1916/2338 [07:28<01:25, 4.95it/s] 82%|████████▏ | 1917/2338 [07:28<01:26, 4.85it/s] 82%|████████▏ | 1918/2338 [07:28<01:33, 4.49it/s] 82%|████████▏ | 1919/2338 [07:29<01:33, 4.49it/s] 82%|████████▏ | 1920/2338 [07:29<01:33, 4.46it/s] 82%|████████▏ | 1921/2338 [07:29<01:32, 4.52it/s] 82%|████████▏ | 1922/2338 [07:29<01:34, 4.42it/s] 82%|████████▏ | 1923/2338 [07:30<01:34, 4.41it/s] 82%|████████▏ | 1924/2338 [07:30<01:38, 4.22it/s] 82%|████████▏ | 1925/2338 [07:30<01:36, 4.27it/s] 82%|████████▏ | 1926/2338 [07:30<01:35, 4.30it/s] 82%|████████▏ | 1927/2338 [07:30<01:33, 4.40it/s] 82%|████████▏ | 1928/2338 [07:31<01:28, 4.63it/s] 83%|████████▎ | 1929/2338 [07:31<01:30, 4.52it/s] 83%|████████▎ | 1930/2338 [07:31<01:30, 4.52it/s] 83%|████████▎ | 1931/2338 [07:31<01:32, 4.42it/s] 83%|████████▎ | 1932/2338 [07:32<01:32, 4.39it/s] 83%|████████▎ | 1933/2338 [07:32<01:32, 4.39it/s] 83%|████████▎ | 1934/2338 [07:32<01:32, 4.39it/s] 83%|████████▎ | 1935/2338 [07:32<01:34, 4.27it/s] 83%|████████▎ | 1936/2338 [07:33<01:42, 3.93it/s] 83%|████████▎ | 1937/2338 [07:33<01:39, 4.04it/s] 83%|████████▎ | 1938/2338 [07:33<01:29, 4.49it/s] 83%|████████▎ | 1939/2338 [07:33<01:32, 4.32it/s] 83%|████████▎ | 1940/2338 [07:33<01:31, 4.37it/s] 83%|████████▎ | 1941/2338 [07:34<01:28, 4.50it/s] 83%|████████▎ | 1942/2338 [07:34<01:25, 4.64it/s] 83%|████████▎ | 1943/2338 [07:34<01:21, 4.86it/s] 83%|████████▎ | 1944/2338 [07:34<01:24, 4.64it/s] 83%|████████▎ | 1945/2338 [07:35<01:26, 4.54it/s] 83%|████████▎ | 1946/2338 [07:35<01:31, 4.28it/s] 83%|████████▎ | 1947/2338 [07:35<01:34, 4.16it/s] 83%|████████▎ | 1948/2338 [07:35<01:32, 4.21it/s] 83%|████████▎ | 1949/2338 [07:36<01:30, 4.29it/s] 83%|████████▎ | 1950/2338 [07:36<01:32, 4.19it/s] 83%|████████▎ | 1951/2338 [07:36<01:32, 4.17it/s] 83%|████████▎ | 1952/2338 [07:36<01:26, 4.47it/s] 84%|████████▎ | 1953/2338 [07:36<01:33, 4.12it/s] 84%|████████▎ | 1954/2338 [07:37<01:33, 4.09it/s] 84%|████████▎ | 1955/2338 [07:37<01:30, 4.22it/s] 84%|████████▎ | 1956/2338 [07:37<01:28, 4.32it/s] 84%|████████▎ | 1957/2338 [07:37<01:25, 4.44it/s] 84%|████████▎ | 1958/2338 [07:38<01:34, 4.04it/s] 84%|████████▍ | 1959/2338 [07:38<01:28, 4.26it/s] 84%|████████▍ | 1960/2338 [07:38<01:25, 4.42it/s] 84%|████████▍ | 1961/2338 [07:38<01:23, 4.50it/s] 84%|████████▍ | 1962/2338 [07:39<01:26, 4.35it/s] 84%|████████▍ | 1963/2338 [07:39<01:25, 4.37it/s] 84%|████████▍ | 1964/2338 [07:39<01:27, 4.28it/s] 84%|████████▍ | 1965/2338 [07:39<01:30, 4.13it/s] 84%|████████▍ | 1966/2338 [07:40<01:38, 3.78it/s] 84%|████████▍ | 1967/2338 [07:40<01:35, 3.90it/s] 84%|████████▍ | 1968/2338 [07:40<01:38, 3.75it/s] 84%|████████▍ | 1969/2338 [07:40<01:38, 3.73it/s] 84%|████████▍ | 1970/2338 [07:41<01:34, 3.91it/s] 84%|████████▍ | 1971/2338 [07:41<01:32, 3.95it/s] 84%|████████▍ | 1972/2338 [07:41<01:21, 4.48it/s] 84%|████████▍ | 1973/2338 [07:41<01:24, 4.31it/s] 84%|████████▍ | 1974/2338 [07:42<01:26, 4.19it/s] 84%|████████▍ | 1975/2338 [07:42<01:26, 4.22it/s] 85%|████████▍ | 1976/2338 [07:42<01:23, 4.33it/s] 85%|████████▍ | 1977/2338 [07:42<01:24, 4.25it/s] 85%|████████▍ | 1978/2338 [07:42<01:27, 4.12it/s] 85%|████████▍ | 1979/2338 [07:43<01:28, 4.06it/s] 85%|████████▍ | 1980/2338 [07:43<01:20, 4.44it/s] 85%|████████▍ | 1981/2338 [07:43<01:24, 4.25it/s] 85%|████████▍ | 1982/2338 [07:43<01:28, 4.04it/s] 85%|████████▍ | 1983/2338 [07:44<01:28, 4.00it/s] 85%|████████▍ | 1984/2338 [07:44<01:26, 4.11it/s] 85%|████████▍ | 1985/2338 [07:44<01:23, 4.21it/s] 85%|████████▍ | 1986/2338 [07:44<01:27, 4.03it/s] 85%|████████▍ | 1987/2338 [07:45<01:31, 3.84it/s] 85%|████████▌ | 1988/2338 [07:45<01:21, 4.30it/s] 85%|████████▌ | 1989/2338 [07:45<01:21, 4.26it/s] 85%|████████▌ | 1990/2338 [07:45<01:22, 4.20it/s] 85%|████████▌ | 1991/2338 [07:46<01:17, 4.47it/s] 85%|████████▌ | 1992/2338 [07:46<01:18, 4.38it/s] 85%|████████▌ | 1993/2338 [07:46<01:14, 4.66it/s] 85%|████████▌ | 1994/2338 [07:46<01:20, 4.28it/s] 85%|████████▌ | 1995/2338 [07:46<01:15, 4.53it/s] 85%|████████▌ | 1996/2338 [07:47<01:18, 4.37it/s] 85%|████████▌ | 1997/2338 [07:47<01:10, 4.83it/s] 85%|████████▌ | 1998/2338 [07:47<01:16, 4.46it/s] 86%|████████▌ | 1999/2338 [07:47<01:19, 4.27it/s] 86%|████████▌ | 2000/2338 [07:48<01:19, 4.26it/s] 86%|████████▌ | 2001/2338 [07:48<01:19, 4.26it/s] 86%|████████▌ | 2002/2338 [07:48<01:20, 4.20it/s] 86%|████████▌ | 2003/2338 [07:48<01:17, 4.33it/s] 86%|████████▌ | 2004/2338 [07:49<01:18, 4.26it/s] 86%|████████▌ | 2005/2338 [07:49<01:21, 4.07it/s] 86%|████████▌ | 2006/2338 [07:49<01:18, 4.22it/s] 86%|████████▌ | 2007/2338 [07:49<01:19, 4.17it/s] 86%|████████▌ | 2008/2338 [07:49<01:14, 4.44it/s] 86%|████████▌ | 2009/2338 [07:50<01:14, 4.39it/s] 86%|████████▌ | 2010/2338 [07:50<01:06, 4.91it/s] 86%|████████▌ | 2011/2338 [07:50<01:10, 4.66it/s] 86%|████████▌ | 2012/2338 [07:50<01:14, 4.36it/s] 86%|████████▌ | 2013/2338 [07:51<01:14, 4.36it/s] 86%|████████▌ | 2014/2338 [07:51<01:14, 4.32it/s] 86%|████████▌ | 2015/2338 [07:51<01:14, 4.31it/s] 86%|████████▌ | 2016/2338 [07:51<01:17, 4.14it/s] 86%|████████▋ | 2017/2338 [07:52<01:18, 4.11it/s] 86%|████████▋ | 2018/2338 [07:52<01:18, 4.07it/s] 86%|████████▋ | 2019/2338 [07:52<01:20, 3.97it/s] 86%|████████▋ | 2020/2338 [07:52<01:17, 4.08it/s] 86%|████████▋ | 2021/2338 [07:53<01:19, 3.99it/s] 86%|████████▋ | 2022/2338 [07:53<01:20, 3.95it/s] 87%|████████▋ | 2023/2338 [07:53<01:17, 4.05it/s] 87%|████████▋ | 2024/2338 [07:53<01:17, 4.03it/s] 87%|████████▋ | 2025/2338 [07:54<01:19, 3.92it/s] 87%|████████▋ | 2026/2338 [07:54<01:16, 4.10it/s] 87%|████████▋ | 2027/2338 [07:54<01:15, 4.13it/s] 87%|████████▋ | 2028/2338 [07:54<01:16, 4.08it/s] 87%|████████▋ | 2029/2338 [07:55<01:12, 4.28it/s] 87%|████████▋ | 2030/2338 [07:55<01:11, 4.29it/s] 87%|████████▋ | 2031/2338 [07:55<01:11, 4.27it/s] 87%|████████▋ | 2032/2338 [07:55<01:11, 4.30it/s] 87%|████████▋ | 2033/2338 [07:55<01:09, 4.40it/s] 87%|████████▋ | 2034/2338 [07:56<01:11, 4.23it/s] 87%|████████▋ | 2035/2338 [07:56<01:12, 4.20it/s] 87%|████████▋ | 2036/2338 [07:56<01:10, 4.28it/s] 87%|████████▋ | 2037/2338 [07:56<01:11, 4.24it/s] 87%|████████▋ | 2038/2338 [07:57<01:03, 4.69it/s] 87%|████████▋ | 2039/2338 [07:57<01:04, 4.61it/s] 87%|████████▋ | 2040/2338 [07:57<01:04, 4.64it/s] 87%|████████▋ | 2041/2338 [07:57<01:06, 4.49it/s] 87%|████████▋ | 2042/2338 [07:57<01:07, 4.38it/s] 87%|████████▋ | 2043/2338 [07:58<01:13, 4.04it/s] 87%|████████▋ | 2044/2338 [07:58<01:14, 3.96it/s] 87%|████████▋ | 2045/2338 [07:58<01:11, 4.09it/s] 88%|████████▊ | 2046/2338 [07:59<01:11, 4.06it/s] 88%|████████▊ | 2047/2338 [07:59<01:10, 4.11it/s] 88%|████████▊ | 2048/2338 [07:59<01:11, 4.08it/s] 88%|████████▊ | 2049/2338 [07:59<01:09, 4.15it/s] 88%|████████▊ | 2050/2338 [07:59<01:10, 4.07it/s] 88%|████████▊ | 2051/2338 [08:00<01:16, 3.77it/s] 88%|████████▊ | 2052/2338 [08:00<01:15, 3.77it/s] 88%|████████▊ | 2053/2338 [08:00<01:10, 4.05it/s] 88%|████████▊ | 2054/2338 [08:01<01:11, 3.95it/s] 88%|████████▊ | 2055/2338 [08:01<01:10, 4.02it/s] 88%|████████▊ | 2056/2338 [08:01<01:08, 4.11it/s] 88%|████████▊ | 2057/2338 [08:01<01:07, 4.15it/s] 88%|████████▊ | 2058/2338 [08:01<01:07, 4.17it/s] 88%|████████▊ | 2059/2338 [08:02<01:10, 3.97it/s] 88%|████████▊ | 2060/2338 [08:02<01:06, 4.18it/s] 88%|████████▊ | 2061/2338 [08:02<01:10, 3.91it/s] 88%|████████▊ | 2062/2338 [08:02<01:07, 4.07it/s] 88%|████████▊ | 2063/2338 [08:03<01:06, 4.11it/s] 88%|████████▊ | 2064/2338 [08:03<01:08, 3.99it/s] 88%|████████▊ | 2065/2338 [08:03<01:07, 4.05it/s] 88%|████████▊ | 2066/2338 [08:03<01:05, 4.13it/s] 88%|████████▊ | 2067/2338 [08:04<01:05, 4.16it/s] 88%|████████▊ | 2068/2338 [08:04<01:04, 4.19it/s] 88%|████████▊ | 2069/2338 [08:04<00:57, 4.68it/s] 89%|████████▊ | 2070/2338 [08:04<00:58, 4.58it/s] 89%|████████▊ | 2071/2338 [08:05<00:59, 4.48it/s] 89%|████████▊ | 2072/2338 [08:05<00:57, 4.60it/s] 89%|████████▊ | 2073/2338 [08:05<00:58, 4.50it/s] 89%|████████▊ | 2074/2338 [08:05<00:59, 4.45it/s] 89%|████████▉ | 2075/2338 [08:05<01:00, 4.35it/s] 89%|████████▉ | 2076/2338 [08:06<01:01, 4.29it/s] 89%|████████▉ | 2077/2338 [08:06<01:01, 4.23it/s] 89%|████████▉ | 2078/2338 [08:06<01:00, 4.28it/s] 89%|████████▉ | 2079/2338 [08:06<01:02, 4.16it/s] 89%|████████▉ | 2080/2338 [08:07<01:01, 4.23it/s] 89%|████████▉ | 2081/2338 [08:07<01:00, 4.24it/s] 89%|████████▉ | 2082/2338 [08:07<01:02, 4.07it/s] 89%|████████▉ | 2083/2338 [08:07<01:00, 4.23it/s] 89%|████████▉ | 2084/2338 [08:08<00:59, 4.29it/s] 89%|████████▉ | 2085/2338 [08:08<00:58, 4.29it/s] 89%|████████▉ | 2086/2338 [08:08<01:00, 4.18it/s] 89%|████████▉ | 2087/2338 [08:08<00:59, 4.25it/s] 89%|████████▉ | 2088/2338 [08:09<00:56, 4.45it/s] 89%|████████▉ | 2089/2338 [08:09<00:54, 4.53it/s] 89%|████████▉ | 2090/2338 [08:09<00:57, 4.32it/s] 89%|████████▉ | 2091/2338 [08:09<00:59, 4.14it/s] 89%|████████▉ | 2092/2338 [08:09<01:00, 4.08it/s] 90%|████████▉ | 2093/2338 [08:10<00:56, 4.36it/s] 90%|████████▉ | 2094/2338 [08:10<00:56, 4.29it/s] 90%|████████▉ | 2095/2338 [08:10<00:55, 4.40it/s] 90%|████████▉ | 2096/2338 [08:10<00:54, 4.43it/s] 90%|████████▉ | 2097/2338 [08:11<00:58, 4.10it/s] 90%|████████▉ | 2098/2338 [08:11<00:58, 4.12it/s] 90%|████████▉ | 2099/2338 [08:11<00:58, 4.09it/s] 90%|████████▉ | 2100/2338 [08:11<00:58, 4.07it/s] 90%|████████▉ | 2101/2338 [08:12<00:56, 4.19it/s] 90%|████████▉ | 2102/2338 [08:12<00:56, 4.19it/s] 90%|████████▉ | 2103/2338 [08:12<00:58, 4.03it/s] 90%|████████▉ | 2104/2338 [08:12<00:58, 4.03it/s] 90%|█████████ | 2105/2338 [08:13<00:57, 4.04it/s] 90%|█████████ | 2106/2338 [08:13<00:58, 4.00it/s] 90%|█████████ | 2107/2338 [08:13<00:57, 4.00it/s] 90%|█████████ | 2108/2338 [08:13<00:58, 3.93it/s] 90%|█████████ | 2109/2338 [08:14<00:56, 4.05it/s] 90%|█████████ | 2110/2338 [08:14<00:56, 4.03it/s] 90%|█████████ | 2111/2338 [08:14<00:54, 4.18it/s] 90%|█████████ | 2112/2338 [08:14<00:55, 4.09it/s] 90%|█████████ | 2113/2338 [08:15<00:52, 4.25it/s] 90%|█████████ | 2114/2338 [08:15<00:51, 4.35it/s] 90%|█████████ | 2115/2338 [08:15<00:52, 4.21it/s] 91%|█████████ | 2116/2338 [08:15<00:54, 4.05it/s] 91%|█████████ | 2117/2338 [08:16<00:54, 4.09it/s] 91%|█████████ | 2118/2338 [08:16<00:54, 4.07it/s] 91%|█████████ | 2119/2338 [08:16<00:53, 4.09it/s] 91%|█████████ | 2120/2338 [08:16<00:50, 4.33it/s] 91%|█████████ | 2121/2338 [08:16<00:47, 4.59it/s] 91%|█████████ | 2122/2338 [08:17<00:46, 4.60it/s] 91%|█████████ | 2123/2338 [08:17<00:49, 4.36it/s] 91%|█████████ | 2124/2338 [08:17<00:48, 4.44it/s] 91%|█████████ | 2125/2338 [08:17<00:51, 4.18it/s] 91%|█████████ | 2126/2338 [08:18<00:50, 4.20it/s] 91%|█████████ | 2127/2338 [08:18<00:53, 3.94it/s] 91%|█████████ | 2128/2338 [08:18<00:53, 3.90it/s] 91%|█████████ | 2129/2338 [08:18<00:54, 3.87it/s] 91%|█████████ | 2130/2338 [08:19<00:52, 3.95it/s] 91%|█████████ | 2131/2338 [08:19<00:54, 3.83it/s] 91%|█████████ | 2132/2338 [08:19<00:49, 4.13it/s] 91%|█████████ | 2133/2338 [08:19<00:50, 4.05it/s] 91%|█████████▏| 2134/2338 [08:20<00:50, 4.04it/s] 91%|█████████▏| 2135/2338 [08:20<00:49, 4.09it/s] 91%|█████████▏| 2136/2338 [08:20<00:49, 4.12it/s] 91%|█████████▏| 2137/2338 [08:20<00:48, 4.14it/s] 91%|█████████▏| 2138/2338 [08:21<00:48, 4.15it/s] 91%|█████████▏| 2139/2338 [08:21<00:49, 4.04it/s] 92%|█████████▏| 2140/2338 [08:21<00:48, 4.10it/s] 92%|█████████▏| 2141/2338 [08:21<00:48, 4.06it/s] 92%|█████████▏| 2142/2338 [08:22<00:46, 4.23it/s] 92%|█████████▏| 2143/2338 [08:22<00:46, 4.23it/s] 92%|█████████▏| 2144/2338 [08:22<00:46, 4.19it/s] 92%|█████████▏| 2145/2338 [08:22<00:45, 4.20it/s] 92%|█████████▏| 2146/2338 [08:23<00:46, 4.11it/s] 92%|█████████▏| 2147/2338 [08:23<00:47, 4.03it/s] 92%|█████████▏| 2148/2338 [08:23<00:45, 4.16it/s] 92%|█████████▏| 2149/2338 [08:23<00:45, 4.13it/s] 92%|█████████▏| 2150/2338 [08:23<00:44, 4.24it/s] 92%|█████████▏| 2151/2338 [08:24<00:43, 4.29it/s] 92%|█████████▏| 2152/2338 [08:24<00:42, 4.36it/s] 92%|█████████▏| 2153/2338 [08:24<00:43, 4.26it/s] 92%|█████████▏| 2154/2338 [08:24<00:42, 4.32it/s] 92%|█████████▏| 2155/2338 [08:25<00:41, 4.37it/s] 92%|█████████▏| 2156/2338 [08:25<00:41, 4.35it/s] 92%|█████████▏| 2157/2338 [08:25<00:41, 4.36it/s] 92%|█████████▏| 2158/2338 [08:25<00:41, 4.31it/s] 92%|█████████▏| 2159/2338 [08:26<00:41, 4.32it/s] 92%|█████████▏| 2160/2338 [08:26<00:41, 4.25it/s] 92%|█████████▏| 2161/2338 [08:26<00:41, 4.28it/s] 92%|█████████▏| 2162/2338 [08:26<00:40, 4.34it/s] 93%|█████████▎| 2163/2338 [08:26<00:40, 4.32it/s] 93%|█████████▎| 2164/2338 [08:27<00:40, 4.33it/s] 93%|█████████▎| 2165/2338 [08:27<00:40, 4.28it/s] 93%|█████████▎| 2166/2338 [08:27<00:38, 4.45it/s] 93%|█████████▎| 2167/2338 [08:27<00:38, 4.44it/s] 93%|█████████▎| 2168/2338 [08:28<00:38, 4.43it/s] 93%|█████████▎| 2169/2338 [08:28<00:40, 4.16it/s] 93%|█████████▎| 2170/2338 [08:28<00:37, 4.46it/s] 93%|█████████▎| 2171/2338 [08:28<00:37, 4.49it/s] 93%|█████████▎| 2172/2338 [08:29<00:37, 4.43it/s] 93%|█████████▎| 2173/2338 [08:29<00:37, 4.41it/s] 93%|█████████▎| 2174/2338 [08:29<00:37, 4.39it/s] 93%|█████████▎| 2175/2338 [08:29<00:38, 4.24it/s] 93%|█████████▎| 2176/2338 [08:29<00:38, 4.16it/s] 93%|█████████▎| 2177/2338 [08:30<00:38, 4.16it/s] 93%|█████████▎| 2178/2338 [08:30<00:38, 4.14it/s] 93%|█████████▎| 2179/2338 [08:30<00:38, 4.18it/s] 93%|█████████▎| 2180/2338 [08:30<00:38, 4.07it/s] 93%|█████████▎| 2181/2338 [08:31<00:38, 4.03it/s] 93%|█████████▎| 2182/2338 [08:31<00:37, 4.14it/s] 93%|█████████▎| 2183/2338 [08:31<00:34, 4.49it/s] 93%|█████████▎| 2184/2338 [08:31<00:34, 4.42it/s] 93%|█████████▎| 2185/2338 [08:32<00:36, 4.19it/s] 93%|█████████▎| 2186/2338 [08:32<00:38, 3.94it/s] 94%|█████████▎| 2187/2338 [08:32<00:37, 4.06it/s] 94%|█████████▎| 2188/2338 [08:32<00:36, 4.12it/s] 94%|█████████▎| 2189/2338 [08:33<00:35, 4.22it/s] 94%|█████████▎| 2190/2338 [08:33<00:36, 4.06it/s] 94%|█████████▎| 2191/2338 [08:33<00:34, 4.31it/s] 94%|█████████▍| 2192/2338 [08:33<00:32, 4.44it/s] 94%|█████████▍| 2193/2338 [08:34<00:35, 4.10it/s] 94%|█████████▍| 2194/2338 [08:34<00:36, 3.96it/s] 94%|█████████▍| 2195/2338 [08:34<00:35, 3.99it/s] 94%|█████████▍| 2196/2338 [08:34<00:34, 4.09it/s] 94%|█████████▍| 2197/2338 [08:35<00:33, 4.15it/s] 94%|█████████▍| 2198/2338 [08:35<00:34, 4.10it/s] 94%|█████████▍| 2199/2338 [08:35<00:34, 4.06it/s] 94%|█████████▍| 2200/2338 [08:35<00:34, 4.06it/s] 94%|█████████▍| 2201/2338 [08:36<00:32, 4.16it/s] 94%|█████████▍| 2202/2338 [08:36<00:30, 4.42it/s] 94%|█████████▍| 2203/2338 [08:36<00:31, 4.30it/s] 94%|█████████▍| 2204/2338 [08:36<00:30, 4.35it/s] 94%|█████████▍| 2205/2338 [08:36<00:31, 4.22it/s] 94%|█████████▍| 2206/2338 [08:37<00:31, 4.18it/s] 94%|█████████▍| 2207/2338 [08:37<00:34, 3.83it/s] 94%|█████████▍| 2208/2338 [08:37<00:32, 4.00it/s] 94%|█████████▍| 2209/2338 [08:37<00:31, 4.11it/s] 95%|█████████▍| 2210/2338 [08:38<00:29, 4.40it/s] 95%|█████████▍| 2211/2338 [08:38<00:30, 4.13it/s] 95%|█████████▍| 2212/2338 [08:38<00:30, 4.17it/s] 95%|█████████▍| 2213/2338 [08:38<00:30, 4.15it/s] 95%|█████████▍| 2214/2338 [08:39<00:27, 4.48it/s] 95%|█████████▍| 2215/2338 [08:39<00:28, 4.33it/s] 95%|█████████▍| 2216/2338 [08:39<00:28, 4.28it/s] 95%|█████████▍| 2217/2338 [08:39<00:27, 4.41it/s] 95%|█████████▍| 2218/2338 [08:39<00:26, 4.53it/s] 95%|█████████▍| 2219/2338 [08:40<00:26, 4.44it/s] 95%|█████████▍| 2220/2338 [08:40<00:26, 4.41it/s] 95%|█████████▍| 2221/2338 [08:40<00:26, 4.47it/s] 95%|█████████▌| 2222/2338 [08:40<00:26, 4.42it/s] 95%|█████████▌| 2223/2338 [08:41<00:25, 4.58it/s] 95%|█████████▌| 2224/2338 [08:41<00:28, 4.03it/s] 95%|█████████▌| 2225/2338 [08:41<00:27, 4.13it/s] 95%|█████████▌| 2226/2338 [08:41<00:26, 4.22it/s] 95%|█████████▌| 2227/2338 [08:42<00:26, 4.20it/s] 95%|█████████▌| 2228/2338 [08:42<00:24, 4.45it/s] 95%|█████████▌| 2229/2338 [08:42<00:25, 4.20it/s] 95%|█████████▌| 2230/2338 [08:42<00:25, 4.30it/s] 95%|█████████▌| 2231/2338 [08:43<00:24, 4.40it/s] 95%|█████████▌| 2232/2338 [08:43<00:24, 4.38it/s] 96%|█████████▌| 2233/2338 [08:43<00:23, 4.38it/s] 96%|█████████▌| 2234/2338 [08:43<00:23, 4.36it/s] 96%|█████████▌| 2235/2338 [08:43<00:23, 4.45it/s] 96%|█████████▌| 2236/2338 [08:44<00:24, 4.22it/s] 96%|█████████▌| 2237/2338 [08:44<00:24, 4.08it/s] 96%|█████████▌| 2238/2338 [08:44<00:24, 4.11it/s] 96%|█████████▌| 2239/2338 [08:44<00:23, 4.16it/s] 96%|█████████▌| 2240/2338 [08:45<00:23, 4.25it/s] 96%|█████████▌| 2241/2338 [08:45<00:23, 4.18it/s] 96%|█████████▌| 2242/2338 [08:45<00:24, 3.97it/s] 96%|█████████▌| 2243/2338 [08:45<00:22, 4.18it/s] 96%|█████████▌| 2244/2338 [08:46<00:20, 4.56it/s] 96%|█████████▌| 2245/2338 [08:46<00:21, 4.29it/s] 96%|█████████▌| 2246/2338 [08:46<00:21, 4.37it/s] 96%|█████████▌| 2247/2338 [08:46<00:20, 4.36it/s] 96%|█████████▌| 2248/2338 [08:46<00:20, 4.37it/s] 96%|█████████▌| 2249/2338 [08:47<00:21, 4.17it/s] 96%|█████████▌| 2250/2338 [08:47<00:20, 4.21it/s] 96%|█████████▋| 2251/2338 [08:47<00:20, 4.22it/s] 96%|█████████▋| 2252/2338 [08:47<00:20, 4.15it/s] 96%|█████████▋| 2253/2338 [08:48<00:19, 4.41it/s] 96%|█████████▋| 2254/2338 [08:48<00:19, 4.42it/s] 96%|█████████▋| 2255/2338 [08:48<00:18, 4.47it/s] 96%|█████████▋| 2256/2338 [08:48<00:17, 4.70it/s] 97%|█████████▋| 2257/2338 [08:49<00:17, 4.60it/s] 97%|█████████▋| 2258/2338 [08:49<00:18, 4.37it/s] 97%|█████████▋| 2259/2338 [08:49<00:18, 4.37it/s] 97%|█████████▋| 2260/2338 [08:49<00:18, 4.18it/s] 97%|█████████▋| 2261/2338 [08:50<00:18, 4.14it/s] 97%|█████████▋| 2262/2338 [08:50<00:17, 4.31it/s] 97%|█████████▋| 2263/2338 [08:50<00:16, 4.56it/s] 97%|█████████▋| 2264/2338 [08:50<00:16, 4.41it/s] 97%|█████████▋| 2265/2338 [08:50<00:16, 4.44it/s] 97%|█████████▋| 2266/2338 [08:51<00:17, 4.14it/s] 97%|█████████▋| 2267/2338 [08:51<00:17, 4.17it/s] 97%|█████████▋| 2268/2338 [08:51<00:16, 4.17it/s] 97%|█████████▋| 2269/2338 [08:51<00:17, 4.03it/s] 97%|█████████▋| 2270/2338 [08:52<00:17, 3.84it/s] 97%|█████████▋| 2271/2338 [08:52<00:17, 3.87it/s] 97%|█████████▋| 2272/2338 [08:52<00:15, 4.22it/s] 97%|█████████▋| 2273/2338 [08:52<00:15, 4.20it/s] 97%|█████████▋| 2274/2338 [08:53<00:15, 4.15it/s] 97%|█████████▋| 2275/2338 [08:53<00:15, 4.09it/s] 97%|█████████▋| 2276/2338 [08:53<00:15, 3.98it/s] 97%|█████████▋| 2277/2338 [08:53<00:14, 4.23it/s] 97%|█████████▋| 2278/2338 [08:54<00:13, 4.55it/s] 97%|█████████▋| 2279/2338 [08:54<00:13, 4.50it/s] 98%|█████████▊| 2280/2338 [08:54<00:13, 4.29it/s] 98%|█████████▊| 2281/2338 [08:54<00:14, 4.03it/s] 98%|█████████▊| 2282/2338 [08:55<00:13, 4.03it/s] 98%|█████████▊| 2283/2338 [08:55<00:13, 4.16it/s] 98%|█████████▊| 2284/2338 [08:55<00:12, 4.21it/s] 98%|█████████▊| 2285/2338 [08:55<00:12, 4.33it/s] 98%|█████████▊| 2286/2338 [08:55<00:12, 4.29it/s] 98%|█████████▊| 2287/2338 [08:56<00:11, 4.35it/s] 98%|█████████▊| 2288/2338 [08:56<00:11, 4.34it/s] 98%|█████████▊| 2289/2338 [08:56<00:11, 4.21it/s] 98%|█████████▊| 2290/2338 [08:56<00:12, 3.92it/s] 98%|█████████▊| 2291/2338 [08:57<00:11, 4.11it/s] 98%|█████████▊| 2292/2338 [08:57<00:11, 4.02it/s] 98%|█████████▊| 2293/2338 [08:57<00:11, 3.99it/s] 98%|█████████▊| 2294/2338 [08:57<00:10, 4.20it/s] 98%|█████████▊| 2295/2338 [08:58<00:10, 4.04it/s] 98%|█████████▊| 2296/2338 [08:58<00:09, 4.44it/s] 98%|█████████▊| 2297/2338 [08:58<00:09, 4.42it/s] 98%|█████████▊| 2298/2338 [08:58<00:08, 4.85it/s] 98%|█████████▊| 2299/2338 [08:58<00:08, 4.76it/s] 98%|█████████▊| 2300/2338 [08:59<00:08, 4.34it/s] 98%|█████████▊| 2301/2338 [08:59<00:08, 4.23it/s] 98%|█████████▊| 2302/2338 [08:59<00:08, 4.29it/s] 99%|█████████▊| 2303/2338 [08:59<00:08, 4.14it/s] 99%|█████████▊| 2304/2338 [09:00<00:07, 4.53it/s] 99%|█████████▊| 2305/2338 [09:00<00:07, 4.41it/s] 99%|█████████▊| 2306/2338 [09:00<00:07, 4.18it/s] 99%|█████████▊| 2307/2338 [09:00<00:07, 4.06it/s] 99%|█████████▊| 2308/2338 [09:01<00:06, 4.29it/s] 99%|█████████▉| 2309/2338 [09:01<00:06, 4.40it/s] 99%|█████████▉| 2310/2338 [09:01<00:06, 4.14it/s] 99%|█████████▉| 2311/2338 [09:01<00:06, 4.06it/s] 99%|█████████▉| 2312/2338 [09:02<00:06, 4.06it/s] 99%|█████████▉| 2313/2338 [09:02<00:06, 3.94it/s] 99%|█████████▉| 2314/2338 [09:02<00:06, 3.89it/s] 99%|█████████▉| 2315/2338 [09:02<00:05, 4.04it/s] 99%|█████████▉| 2316/2338 [09:03<00:05, 4.09it/s] 99%|█████████▉| 2317/2338 [09:03<00:05, 4.19it/s] 99%|█████████▉| 2318/2338 [09:03<00:04, 4.19it/s] 99%|█████████▉| 2319/2338 [09:03<00:04, 4.28it/s] 99%|█████████▉| 2320/2338 [09:04<00:04, 4.30it/s] 99%|█████████▉| 2321/2338 [09:04<00:03, 4.25it/s] 99%|█████████▉| 2322/2338 [09:04<00:03, 4.37it/s] 99%|█████████▉| 2323/2338 [09:04<00:03, 4.30it/s] 99%|█████████▉| 2324/2338 [09:04<00:03, 4.34it/s] 99%|█████████▉| 2325/2338 [09:05<00:02, 4.48it/s] 99%|█████████▉| 2326/2338 [09:05<00:02, 4.43it/s] 100%|█████████▉| 2327/2338 [09:05<00:02, 4.34it/s] 100%|█████████▉| 2328/2338 [09:05<00:02, 4.41it/s] 100%|█████████▉| 2329/2338 [09:06<00:02, 4.19it/s] 100%|█████████▉| 2330/2338 [09:06<00:01, 4.19it/s] 100%|█████████▉| 2331/2338 [09:06<00:01, 4.32it/s] 100%|█████████▉| 2332/2338 [09:06<00:01, 4.29it/s] 100%|█████████▉| 2333/2338 [09:07<00:01, 4.32it/s] 100%|█████████▉| 2334/2338 [09:07<00:00, 4.11it/s] 100%|█████████▉| 2335/2338 [09:07<00:00, 4.27it/s] 100%|█████████▉| 2336/2338 [09:07<00:00, 4.36it/s] 100%|█████████▉| 2337/2338 [09:07<00:00, 4.54it/s] 100%|██████████| 2338/2338 [09:08<00:00, 4.37it/s] 100%|██████████| 2338/2338 [09:08<00:00, 4.27it/s]
_generate_tf_records(valid_prefix,
_URLS['validation_images'],
valid_annotation_path)
0%| | 0/422 [00:00<?, ?it/s] 0%| | 1/422 [00:00<01:10, 5.99it/s] 0%| | 2/422 [00:00<01:12, 5.78it/s] 1%| | 3/422 [00:00<01:13, 5.70it/s] 1%| | 4/422 [00:00<01:12, 5.73it/s] 1%| | 5/422 [00:00<01:22, 5.07it/s] 1%|▏ | 6/422 [00:01<01:22, 5.03it/s] 2%|▏ | 7/422 [00:01<01:16, 5.45it/s] 2%|▏ | 8/422 [00:01<01:19, 5.22it/s] 2%|▏ | 9/422 [00:01<01:17, 5.30it/s] 2%|▏ | 10/422 [00:01<01:15, 5.45it/s] 3%|▎ | 11/422 [00:02<01:19, 5.14it/s] 3%|▎ | 12/422 [00:02<01:15, 5.44it/s] 3%|▎ | 13/422 [00:02<01:11, 5.71it/s] 3%|▎ | 14/422 [00:02<01:12, 5.60it/s] 4%|▎ | 15/422 [00:02<01:17, 5.22it/s] 4%|▍ | 16/422 [00:02<01:19, 5.11it/s] 4%|▍ | 17/422 [00:03<01:18, 5.19it/s] 4%|▍ | 18/422 [00:03<01:14, 5.42it/s] 5%|▍ | 19/422 [00:03<01:16, 5.30it/s] 5%|▍ | 20/422 [00:03<01:17, 5.16it/s] 5%|▍ | 21/422 [00:03<01:13, 5.43it/s] 5%|▌ | 22/422 [00:04<01:17, 5.17it/s] 5%|▌ | 23/422 [00:04<01:11, 5.56it/s] 6%|▌ | 24/422 [00:04<01:17, 5.15it/s] 6%|▌ | 25/422 [00:04<01:19, 5.01it/s] 6%|▌ | 26/422 [00:04<01:26, 4.59it/s] 6%|▋ | 27/422 [00:05<01:23, 4.71it/s] 7%|▋ | 28/422 [00:05<01:22, 4.80it/s] 7%|▋ | 29/422 [00:05<01:19, 4.92it/s] 7%|▋ | 30/422 [00:05<01:21, 4.83it/s] 7%|▋ | 31/422 [00:06<01:25, 4.60it/s] 8%|▊ | 32/422 [00:06<01:28, 4.41it/s] 8%|▊ | 33/422 [00:06<01:24, 4.63it/s] 8%|▊ | 34/422 [00:06<01:18, 4.91it/s] 8%|▊ | 35/422 [00:06<01:15, 5.15it/s] 9%|▊ | 36/422 [00:06<01:12, 5.32it/s] 9%|▉ | 37/422 [00:07<01:12, 5.29it/s] 9%|▉ | 38/422 [00:07<01:12, 5.27it/s] 9%|▉ | 39/422 [00:07<01:18, 4.89it/s] 9%|▉ | 40/422 [00:07<01:13, 5.22it/s] 10%|▉ | 41/422 [00:07<01:11, 5.36it/s] 10%|▉ | 42/422 [00:08<01:12, 5.22it/s] 10%|█ | 43/422 [00:08<01:18, 4.80it/s] 10%|█ | 44/422 [00:08<01:13, 5.17it/s] 11%|█ | 45/422 [00:08<01:19, 4.72it/s] 11%|█ | 46/422 [00:08<01:14, 5.02it/s] 11%|█ | 47/422 [00:09<01:10, 5.31it/s] 11%|█▏ | 48/422 [00:09<01:09, 5.41it/s] 12%|█▏ | 49/422 [00:09<01:09, 5.36it/s] 12%|█▏ | 50/422 [00:09<01:07, 5.49it/s] 12%|█▏ | 51/422 [00:09<01:06, 5.54it/s] 12%|█▏ | 52/422 [00:10<01:06, 5.59it/s] 13%|█▎ | 53/422 [00:10<01:04, 5.71it/s] 13%|█▎ | 54/422 [00:10<01:12, 5.07it/s] 13%|█▎ | 55/422 [00:10<01:09, 5.27it/s] 13%|█▎ | 56/422 [00:10<01:07, 5.39it/s] 14%|█▎ | 57/422 [00:11<01:11, 5.09it/s] 14%|█▎ | 58/422 [00:11<01:09, 5.20it/s] 14%|█▍ | 59/422 [00:11<01:17, 4.66it/s] 14%|█▍ | 60/422 [00:11<01:16, 4.73it/s] 14%|█▍ | 61/422 [00:11<01:21, 4.44it/s] 15%|█▍ | 62/422 [00:12<01:14, 4.82it/s] 15%|█▍ | 63/422 [00:12<01:12, 4.93it/s] 15%|█▌ | 64/422 [00:12<01:16, 4.70it/s] 15%|█▌ | 65/422 [00:12<01:08, 5.18it/s] 16%|█▌ | 66/422 [00:12<01:03, 5.62it/s] 16%|█▌ | 67/422 [00:12<01:02, 5.67it/s] 16%|█▌ | 68/422 [00:13<01:13, 4.82it/s] 16%|█▋ | 69/422 [00:13<01:09, 5.06it/s] 17%|█▋ | 70/422 [00:13<01:13, 4.77it/s] 17%|█▋ | 71/422 [00:13<01:14, 4.74it/s] 17%|█▋ | 72/422 [00:14<01:14, 4.73it/s] 17%|█▋ | 73/422 [00:14<01:13, 4.74it/s] 18%|█▊ | 74/422 [00:14<01:09, 5.04it/s] 18%|█▊ | 75/422 [00:14<01:11, 4.88it/s] 18%|█▊ | 76/422 [00:14<01:12, 4.79it/s] 18%|█▊ | 77/422 [00:15<01:07, 5.08it/s] 18%|█▊ | 78/422 [00:15<01:10, 4.89it/s] 19%|█▊ | 79/422 [00:15<01:12, 4.75it/s] 19%|█▉ | 80/422 [00:15<01:07, 5.05it/s] 19%|█▉ | 81/422 [00:15<01:01, 5.52it/s] 19%|█▉ | 82/422 [00:16<01:03, 5.35it/s] 20%|█▉ | 83/422 [00:16<01:01, 5.54it/s] 20%|█▉ | 84/422 [00:16<00:58, 5.76it/s] 20%|██ | 85/422 [00:16<01:00, 5.56it/s] 20%|██ | 86/422 [00:16<01:00, 5.56it/s] 21%|██ | 87/422 [00:16<01:03, 5.30it/s] 21%|██ | 88/422 [00:17<00:58, 5.74it/s] 21%|██ | 89/422 [00:17<01:00, 5.52it/s] 21%|██▏ | 90/422 [00:17<00:58, 5.66it/s] 22%|██▏ | 91/422 [00:17<01:03, 5.22it/s] 22%|██▏ | 92/422 [00:17<00:59, 5.59it/s] 22%|██▏ | 93/422 [00:17<00:55, 5.93it/s] 22%|██▏ | 94/422 [00:18<00:54, 5.98it/s] 23%|██▎ | 95/422 [00:18<00:57, 5.71it/s] 23%|██▎ | 96/422 [00:18<01:04, 5.04it/s] 23%|██▎ | 97/422 [00:18<01:06, 4.92it/s] 23%|██▎ | 98/422 [00:18<01:04, 5.05it/s] 23%|██▎ | 99/422 [00:19<01:02, 5.16it/s] 24%|██▎ | 100/422 [00:19<01:00, 5.34it/s] 24%|██▍ | 101/422 [00:19<00:59, 5.39it/s] 24%|██▍ | 102/422 [00:19<00:59, 5.38it/s] 24%|██▍ | 103/422 [00:19<00:55, 5.79it/s] 25%|██▍ | 104/422 [00:20<01:00, 5.25it/s] 25%|██▍ | 105/422 [00:20<00:57, 5.50it/s] 25%|██▌ | 106/422 [00:20<00:53, 5.91it/s] 25%|██▌ | 107/422 [00:20<01:00, 5.18it/s] 26%|██▌ | 108/422 [00:20<01:01, 5.12it/s] 26%|██▌ | 109/422 [00:21<01:06, 4.72it/s] 26%|██▌ | 110/422 [00:21<01:08, 4.58it/s] 26%|██▋ | 111/422 [00:21<01:06, 4.65it/s] 27%|██▋ | 112/422 [00:21<00:59, 5.20it/s] 27%|██▋ | 113/422 [00:21<01:02, 4.96it/s] 27%|██▋ | 114/422 [00:22<00:59, 5.15it/s] 27%|██▋ | 115/422 [00:22<01:03, 4.83it/s] 27%|██▋ | 116/422 [00:22<01:06, 4.60it/s] 28%|██▊ | 117/422 [00:22<01:11, 4.25it/s] 28%|██▊ | 118/422 [00:23<01:09, 4.37it/s] 28%|██▊ | 119/422 [00:23<01:10, 4.32it/s] 28%|██▊ | 120/422 [00:23<01:16, 3.95it/s] 29%|██▊ | 121/422 [00:23<01:14, 4.06it/s] 29%|██▉ | 122/422 [00:24<01:09, 4.34it/s] 29%|██▉ | 123/422 [00:24<01:09, 4.30it/s] 29%|██▉ | 124/422 [00:24<01:09, 4.27it/s] 30%|██▉ | 125/422 [00:24<01:07, 4.43it/s] 30%|██▉ | 126/422 [00:24<01:06, 4.45it/s] 30%|███ | 127/422 [00:25<01:08, 4.33it/s] 30%|███ | 128/422 [00:25<01:09, 4.26it/s] 31%|███ | 129/422 [00:25<01:09, 4.24it/s] 31%|███ | 130/422 [00:25<01:11, 4.08it/s] 31%|███ | 131/422 [00:26<01:11, 4.05it/s] 31%|███▏ | 132/422 [00:26<01:06, 4.34it/s] 32%|███▏ | 133/422 [00:26<01:10, 4.11it/s] 32%|███▏ | 134/422 [00:26<01:05, 4.39it/s] 32%|███▏ | 135/422 [00:27<01:07, 4.24it/s] 32%|███▏ | 136/422 [00:27<01:09, 4.10it/s] 32%|███▏ | 137/422 [00:27<01:07, 4.21it/s] 33%|███▎ | 138/422 [00:27<01:08, 4.16it/s] 33%|███▎ | 139/422 [00:28<01:07, 4.18it/s] 33%|███▎ | 140/422 [00:28<01:05, 4.29it/s] 33%|███▎ | 141/422 [00:28<01:08, 4.08it/s] 34%|███▎ | 142/422 [00:28<00:59, 4.73it/s] 34%|███▍ | 143/422 [00:28<01:00, 4.64it/s] 34%|███▍ | 144/422 [00:29<01:07, 4.14it/s] 34%|███▍ | 145/422 [00:29<01:07, 4.12it/s] 35%|███▍ | 146/422 [00:29<01:02, 4.43it/s] 35%|███▍ | 147/422 [00:29<01:04, 4.28it/s] 35%|███▌ | 148/422 [00:30<01:01, 4.44it/s] 35%|███▌ | 149/422 [00:30<00:58, 4.68it/s] 36%|███▌ | 150/422 [00:30<01:02, 4.35it/s] 36%|███▌ | 151/422 [00:30<01:03, 4.28it/s] 36%|███▌ | 152/422 [00:30<01:02, 4.35it/s] 36%|███▋ | 153/422 [00:31<01:02, 4.29it/s] 36%|███▋ | 154/422 [00:31<01:01, 4.35it/s] 37%|███▋ | 155/422 [00:31<01:01, 4.36it/s] 37%|███▋ | 156/422 [00:31<00:59, 4.48it/s] 37%|███▋ | 157/422 [00:32<00:58, 4.52it/s] 37%|███▋ | 158/422 [00:32<00:58, 4.50it/s] 38%|███▊ | 159/422 [00:32<01:00, 4.36it/s] 38%|███▊ | 160/422 [00:32<00:57, 4.55it/s] 38%|███▊ | 161/422 [00:33<00:58, 4.45it/s] 38%|███▊ | 162/422 [00:33<00:57, 4.49it/s] 39%|███▊ | 163/422 [00:33<00:58, 4.40it/s] 39%|███▉ | 164/422 [00:33<01:00, 4.28it/s] 39%|███▉ | 165/422 [00:33<00:58, 4.38it/s] 39%|███▉ | 166/422 [00:34<01:00, 4.25it/s] 40%|███▉ | 167/422 [00:34<01:00, 4.19it/s] 40%|███▉ | 168/422 [00:34<01:00, 4.17it/s] 40%|████ | 169/422 [00:34<00:55, 4.53it/s] 40%|████ | 170/422 [00:35<00:57, 4.38it/s] 41%|████ | 171/422 [00:35<00:53, 4.67it/s] 41%|████ | 172/422 [00:35<00:53, 4.66it/s] 41%|████ | 173/422 [00:35<00:56, 4.40it/s] 41%|████ | 174/422 [00:35<00:56, 4.40it/s] 41%|████▏ | 175/422 [00:36<00:57, 4.29it/s] 42%|████▏ | 176/422 [00:36<00:58, 4.23it/s] 42%|████▏ | 177/422 [00:36<01:00, 4.05it/s] 42%|████▏ | 178/422 [00:36<00:58, 4.14it/s] 42%|████▏ | 179/422 [00:37<00:57, 4.21it/s] 43%|████▎ | 180/422 [00:37<00:58, 4.16it/s] 43%|████▎ | 181/422 [00:37<00:55, 4.33it/s] 43%|████▎ | 182/422 [00:37<00:54, 4.38it/s] 43%|████▎ | 183/422 [00:38<00:54, 4.39it/s] 44%|████▎ | 184/422 [00:38<00:55, 4.27it/s] 44%|████▍ | 185/422 [00:38<00:54, 4.33it/s] 44%|████▍ | 186/422 [00:38<00:55, 4.24it/s] 44%|████▍ | 187/422 [00:38<00:50, 4.65it/s] 45%|████▍ | 188/422 [00:39<00:50, 4.60it/s] 45%|████▍ | 189/422 [00:39<00:52, 4.40it/s] 45%|████▌ | 190/422 [00:39<00:53, 4.37it/s] 45%|████▌ | 191/422 [00:39<00:50, 4.56it/s] 45%|████▌ | 192/422 [00:40<00:49, 4.67it/s] 46%|████▌ | 193/422 [00:40<00:52, 4.34it/s] 46%|████▌ | 194/422 [00:40<00:55, 4.13it/s] 46%|████▌ | 195/422 [00:40<00:54, 4.13it/s] 46%|████▋ | 196/422 [00:41<00:53, 4.21it/s] 47%|████▋ | 197/422 [00:41<00:53, 4.23it/s] 47%|████▋ | 198/422 [00:41<00:54, 4.12it/s] 47%|████▋ | 199/422 [00:41<00:52, 4.24it/s] 47%|████▋ | 200/422 [00:42<00:52, 4.23it/s] 48%|████▊ | 201/422 [00:42<00:51, 4.27it/s] 48%|████▊ | 202/422 [00:42<00:52, 4.17it/s] 48%|████▊ | 203/422 [00:42<00:54, 4.03it/s] 48%|████▊ | 204/422 [00:42<00:50, 4.32it/s] 49%|████▊ | 205/422 [00:43<00:50, 4.31it/s] 49%|████▉ | 206/422 [00:43<00:50, 4.29it/s] 49%|████▉ | 207/422 [00:43<00:48, 4.40it/s] 49%|████▉ | 208/422 [00:43<00:48, 4.45it/s] 50%|████▉ | 209/422 [00:44<00:48, 4.37it/s] 50%|████▉ | 210/422 [00:44<00:48, 4.38it/s] 50%|█████ | 211/422 [00:44<00:45, 4.66it/s] 50%|█████ | 212/422 [00:44<00:46, 4.52it/s] 50%|█████ | 213/422 [00:45<00:48, 4.34it/s] 51%|█████ | 214/422 [00:45<00:44, 4.68it/s] 51%|█████ | 215/422 [00:45<00:46, 4.49it/s] 51%|█████ | 216/422 [00:45<00:41, 4.95it/s] 51%|█████▏ | 217/422 [00:45<00:43, 4.75it/s] 52%|█████▏ | 218/422 [00:46<00:42, 4.79it/s] 52%|█████▏ | 219/422 [00:46<00:42, 4.73it/s] 52%|█████▏ | 220/422 [00:46<00:44, 4.58it/s] 52%|█████▏ | 221/422 [00:46<00:42, 4.68it/s] 53%|█████▎ | 222/422 [00:46<00:46, 4.32it/s] 53%|█████▎ | 223/422 [00:47<00:45, 4.35it/s] 53%|█████▎ | 224/422 [00:47<00:44, 4.47it/s] 53%|█████▎ | 225/422 [00:47<00:42, 4.66it/s] 54%|█████▎ | 226/422 [00:47<00:44, 4.39it/s] 54%|█████▍ | 227/422 [00:48<00:40, 4.76it/s] 54%|█████▍ | 228/422 [00:48<00:40, 4.74it/s] 54%|█████▍ | 229/422 [00:48<00:40, 4.78it/s] 55%|█████▍ | 230/422 [00:48<00:40, 4.69it/s] 55%|█████▍ | 231/422 [00:48<00:40, 4.68it/s] 55%|█████▍ | 232/422 [00:49<00:40, 4.74it/s] 55%|█████▌ | 233/422 [00:49<00:40, 4.70it/s] 55%|█████▌ | 234/422 [00:49<00:40, 4.61it/s] 56%|█████▌ | 235/422 [00:49<00:38, 4.80it/s] 56%|█████▌ | 236/422 [00:49<00:39, 4.73it/s] 56%|█████▌ | 237/422 [00:50<00:36, 5.05it/s] 56%|█████▋ | 238/422 [00:50<00:35, 5.16it/s] 57%|█████▋ | 239/422 [00:50<00:37, 4.88it/s] 57%|█████▋ | 240/422 [00:50<00:38, 4.78it/s] 57%|█████▋ | 241/422 [00:50<00:36, 4.94it/s] 57%|█████▋ | 242/422 [00:51<00:36, 4.99it/s] 58%|█████▊ | 243/422 [00:51<00:36, 4.88it/s] 58%|█████▊ | 244/422 [00:51<00:37, 4.75it/s] 58%|█████▊ | 245/422 [00:51<00:37, 4.68it/s] 58%|█████▊ | 246/422 [00:52<00:40, 4.39it/s] 59%|█████▊ | 247/422 [00:52<00:40, 4.34it/s] 59%|█████▉ | 248/422 [00:52<00:37, 4.69it/s] 59%|█████▉ | 249/422 [00:52<00:34, 4.99it/s] 59%|█████▉ | 250/422 [00:52<00:36, 4.66it/s] 59%|█████▉ | 251/422 [00:53<00:40, 4.25it/s] 60%|█████▉ | 252/422 [00:53<00:40, 4.20it/s] 60%|█████▉ | 253/422 [00:53<00:39, 4.28it/s] 60%|██████ | 254/422 [00:53<00:40, 4.17it/s] 60%|██████ | 255/422 [00:54<00:38, 4.33it/s] 61%|██████ | 256/422 [00:54<00:38, 4.34it/s] 61%|██████ | 257/422 [00:54<00:32, 5.06it/s] 61%|██████ | 258/422 [00:54<00:33, 4.95it/s] 61%|██████▏ | 259/422 [00:54<00:35, 4.60it/s] 62%|██████▏ | 260/422 [00:55<00:34, 4.72it/s] 62%|██████▏ | 261/422 [00:55<00:34, 4.66it/s] 62%|██████▏ | 262/422 [00:55<00:35, 4.46it/s] 62%|██████▏ | 263/422 [00:55<00:31, 4.98it/s] 63%|██████▎ | 264/422 [00:55<00:30, 5.16it/s] 63%|██████▎ | 265/422 [00:56<00:30, 5.19it/s] 63%|██████▎ | 266/422 [00:56<00:32, 4.74it/s] 63%|██████▎ | 267/422 [00:56<00:33, 4.63it/s] 64%|██████▎ | 268/422 [00:56<00:33, 4.55it/s] 64%|██████▎ | 269/422 [00:56<00:32, 4.69it/s] 64%|██████▍ | 270/422 [00:57<00:33, 4.57it/s] 64%|██████▍ | 271/422 [00:57<00:33, 4.55it/s] 64%|██████▍ | 272/422 [00:57<00:33, 4.43it/s] 65%|██████▍ | 273/422 [00:57<00:34, 4.29it/s] 65%|██████▍ | 274/422 [00:58<00:33, 4.37it/s] 65%|██████▌ | 275/422 [00:58<00:30, 4.82it/s] 65%|██████▌ | 276/422 [00:58<00:29, 4.96it/s] 66%|██████▌ | 277/422 [00:58<00:33, 4.32it/s] 66%|██████▌ | 278/422 [00:59<00:33, 4.31it/s] 66%|██████▌ | 279/422 [00:59<00:32, 4.38it/s] 66%|██████▋ | 280/422 [00:59<00:32, 4.37it/s] 67%|██████▋ | 281/422 [00:59<00:32, 4.37it/s] 67%|██████▋ | 282/422 [00:59<00:31, 4.39it/s] 67%|██████▋ | 283/422 [01:00<00:31, 4.43it/s] 67%|██████▋ | 284/422 [01:00<00:31, 4.37it/s] 68%|██████▊ | 285/422 [01:00<00:31, 4.38it/s] 68%|██████▊ | 286/422 [01:00<00:31, 4.25it/s] 68%|██████▊ | 287/422 [01:01<00:30, 4.44it/s] 68%|██████▊ | 288/422 [01:01<00:29, 4.48it/s] 68%|██████▊ | 289/422 [01:01<00:30, 4.40it/s] 69%|██████▊ | 290/422 [01:01<00:34, 3.87it/s] 69%|██████▉ | 291/422 [01:02<00:33, 3.94it/s] 69%|██████▉ | 292/422 [01:02<00:32, 3.96it/s] 69%|██████▉ | 293/422 [01:02<00:31, 4.04it/s] 70%|██████▉ | 294/422 [01:02<00:29, 4.41it/s] 70%|██████▉ | 295/422 [01:02<00:27, 4.67it/s] 70%|███████ | 296/422 [01:03<00:27, 4.54it/s] 70%|███████ | 297/422 [01:03<00:28, 4.45it/s] 71%|███████ | 298/422 [01:03<00:27, 4.55it/s] 71%|███████ | 299/422 [01:03<00:27, 4.50it/s] 71%|███████ | 300/422 [01:04<00:29, 4.18it/s] 71%|███████▏ | 301/422 [01:04<00:28, 4.18it/s] 72%|███████▏ | 302/422 [01:04<00:28, 4.19it/s] 72%|███████▏ | 303/422 [01:04<00:28, 4.18it/s] 72%|███████▏ | 304/422 [01:05<00:26, 4.46it/s] 72%|███████▏ | 305/422 [01:05<00:27, 4.19it/s] 73%|███████▎ | 306/422 [01:05<00:26, 4.37it/s] 73%|███████▎ | 307/422 [01:05<00:27, 4.25it/s] 73%|███████▎ | 308/422 [01:05<00:26, 4.25it/s] 73%|███████▎ | 309/422 [01:06<00:26, 4.25it/s] 73%|███████▎ | 310/422 [01:06<00:26, 4.15it/s] 74%|███████▎ | 311/422 [01:06<00:27, 4.07it/s] 74%|███████▍ | 312/422 [01:06<00:27, 4.06it/s] 74%|███████▍ | 313/422 [01:07<00:25, 4.26it/s] 74%|███████▍ | 314/422 [01:07<00:24, 4.42it/s] 75%|███████▍ | 315/422 [01:07<00:24, 4.32it/s] 75%|███████▍ | 316/422 [01:07<00:22, 4.75it/s] 75%|███████▌ | 317/422 [01:08<00:23, 4.56it/s] 75%|███████▌ | 318/422 [01:08<00:22, 4.58it/s] 76%|███████▌ | 319/422 [01:08<00:22, 4.58it/s] 76%|███████▌ | 320/422 [01:08<00:23, 4.41it/s] 76%|███████▌ | 321/422 [01:08<00:23, 4.31it/s] 76%|███████▋ | 322/422 [01:09<00:23, 4.19it/s] 77%|███████▋ | 323/422 [01:09<00:24, 4.06it/s] 77%|███████▋ | 324/422 [01:09<00:22, 4.36it/s] 77%|███████▋ | 325/422 [01:09<00:22, 4.35it/s] 77%|███████▋ | 326/422 [01:10<00:22, 4.23it/s] 77%|███████▋ | 327/422 [01:10<00:23, 4.01it/s] 78%|███████▊ | 328/422 [01:10<00:20, 4.48it/s] 78%|███████▊ | 329/422 [01:10<00:21, 4.30it/s] 78%|███████▊ | 330/422 [01:11<00:20, 4.43it/s] 78%|███████▊ | 331/422 [01:11<00:20, 4.53it/s] 79%|███████▊ | 332/422 [01:11<00:20, 4.36it/s] 79%|███████▉ | 333/422 [01:11<00:20, 4.30it/s] 79%|███████▉ | 334/422 [01:11<00:19, 4.42it/s] 79%|███████▉ | 335/422 [01:12<00:19, 4.44it/s] 80%|███████▉ | 336/422 [01:12<00:19, 4.33it/s] 80%|███████▉ | 337/422 [01:12<00:19, 4.29it/s] 80%|████████ | 338/422 [01:12<00:19, 4.22it/s] 80%|████████ | 339/422 [01:13<00:20, 4.03it/s] 81%|████████ | 340/422 [01:13<00:19, 4.26it/s] 81%|████████ | 341/422 [01:13<00:19, 4.08it/s] 81%|████████ | 342/422 [01:13<00:19, 4.21it/s] 81%|████████▏ | 343/422 [01:14<00:19, 4.11it/s] 82%|████████▏ | 344/422 [01:14<00:19, 4.10it/s] 82%|████████▏ | 345/422 [01:14<00:18, 4.08it/s] 82%|████████▏ | 346/422 [01:14<00:18, 4.05it/s] 82%|████████▏ | 347/422 [01:15<00:18, 4.12it/s] 82%|████████▏ | 348/422 [01:15<00:17, 4.21it/s] 83%|████████▎ | 349/422 [01:15<00:17, 4.17it/s] 83%|████████▎ | 350/422 [01:15<00:16, 4.49it/s] 83%|████████▎ | 351/422 [01:16<00:16, 4.29it/s] 83%|████████▎ | 352/422 [01:16<00:16, 4.35it/s] 84%|████████▎ | 353/422 [01:16<00:15, 4.53it/s] 84%|████████▍ | 354/422 [01:16<00:15, 4.46it/s] 84%|████████▍ | 355/422 [01:16<00:14, 4.47it/s] 84%|████████▍ | 356/422 [01:17<00:14, 4.46it/s] 85%|████████▍ | 357/422 [01:17<00:14, 4.57it/s] 85%|████████▍ | 358/422 [01:17<00:14, 4.57it/s] 85%|████████▌ | 359/422 [01:17<00:13, 4.57it/s] 85%|████████▌ | 360/422 [01:17<00:13, 4.67it/s] 86%|████████▌ | 361/422 [01:18<00:13, 4.56it/s] 86%|████████▌ | 362/422 [01:18<00:14, 4.28it/s] 86%|████████▌ | 363/422 [01:18<00:14, 4.16it/s] 86%|████████▋ | 364/422 [01:19<00:14, 4.07it/s] 86%|████████▋ | 365/422 [01:19<00:13, 4.10it/s] 87%|████████▋ | 366/422 [01:19<00:12, 4.33it/s] 87%|████████▋ | 367/422 [01:19<00:12, 4.35it/s] 87%|████████▋ | 368/422 [01:19<00:12, 4.24it/s] 87%|████████▋ | 369/422 [01:20<00:12, 4.26it/s] 88%|████████▊ | 370/422 [01:20<00:12, 4.32it/s] 88%|████████▊ | 371/422 [01:20<00:12, 4.15it/s] 88%|████████▊ | 372/422 [01:20<00:11, 4.40it/s] 88%|████████▊ | 373/422 [01:21<00:11, 4.44it/s] 89%|████████▊ | 374/422 [01:21<00:11, 4.21it/s] 89%|████████▉ | 375/422 [01:21<00:11, 4.23it/s] 89%|████████▉ | 376/422 [01:21<00:11, 4.06it/s] 89%|████████▉ | 377/422 [01:22<00:10, 4.18it/s] 90%|████████▉ | 378/422 [01:22<00:10, 4.36it/s] 90%|████████▉ | 379/422 [01:22<00:10, 4.06it/s] 90%|█████████ | 380/422 [01:22<00:10, 4.07it/s] 90%|█████████ | 381/422 [01:22<00:08, 4.58it/s] 91%|█████████ | 382/422 [01:23<00:08, 4.54it/s] 91%|█████████ | 383/422 [01:23<00:09, 4.22it/s] 91%|█████████ | 384/422 [01:23<00:09, 4.20it/s] 91%|█████████ | 385/422 [01:23<00:08, 4.35it/s] 91%|█████████▏| 386/422 [01:24<00:08, 4.09it/s] 92%|█████████▏| 387/422 [01:24<00:08, 4.12it/s] 92%|█████████▏| 388/422 [01:24<00:08, 4.04it/s] 92%|█████████▏| 389/422 [01:24<00:08, 4.11it/s] 92%|█████████▏| 390/422 [01:25<00:07, 4.02it/s] 93%|█████████▎| 391/422 [01:25<00:07, 4.13it/s] 93%|█████████▎| 392/422 [01:25<00:07, 4.19it/s] 93%|█████████▎| 393/422 [01:25<00:07, 3.99it/s] 93%|█████████▎| 394/422 [01:26<00:06, 4.03it/s] 94%|█████████▎| 395/422 [01:26<00:06, 4.02it/s] 94%|█████████▍| 396/422 [01:26<00:06, 4.04it/s] 94%|█████████▍| 397/422 [01:26<00:05, 4.33it/s] 94%|█████████▍| 398/422 [01:27<00:05, 4.18it/s] 95%|█████████▍| 399/422 [01:27<00:05, 4.15it/s] 95%|█████████▍| 400/422 [01:27<00:05, 4.24it/s] 95%|█████████▌| 401/422 [01:27<00:04, 4.24it/s] 95%|█████████▌| 402/422 [01:28<00:04, 4.10it/s] 95%|█████████▌| 403/422 [01:28<00:04, 4.51it/s] 96%|█████████▌| 404/422 [01:28<00:04, 4.16it/s] 96%|█████████▌| 405/422 [01:28<00:04, 4.09it/s] 96%|█████████▌| 406/422 [01:28<00:03, 4.25it/s] 96%|█████████▋| 407/422 [01:29<00:03, 4.24it/s] 97%|█████████▋| 408/422 [01:29<00:03, 4.52it/s] 97%|█████████▋| 409/422 [01:29<00:02, 4.55it/s] 97%|█████████▋| 410/422 [01:29<00:02, 4.64it/s] 97%|█████████▋| 411/422 [01:30<00:02, 4.52it/s] 98%|█████████▊| 412/422 [01:30<00:02, 4.47it/s] 98%|█████████▊| 413/422 [01:30<00:02, 4.38it/s] 98%|█████████▊| 414/422 [01:30<00:01, 4.67it/s] 98%|█████████▊| 415/422 [01:30<00:01, 4.36it/s] 99%|█████████▊| 416/422 [01:31<00:01, 4.30it/s] 99%|█████████▉| 417/422 [01:31<00:01, 4.44it/s] 99%|█████████▉| 418/422 [01:31<00:00, 4.61it/s] 99%|█████████▉| 419/422 [01:31<00:00, 4.46it/s] 100%|█████████▉| 420/422 [01:32<00:00, 4.39it/s] 100%|█████████▉| 421/422 [01:32<00:00, 4.54it/s] 100%|██████████| 422/422 [01:32<00:00, 4.20it/s] 100%|██████████| 422/422 [01:32<00:00, 4.56it/s]
Configure the MaskRCNN Resnet FPN COCO model for custom dataset
train_data_input_path = './lvis_tfrecords/train*'
valid_data_input_path = './lvis_tfrecords/val*'
test_data_input_path = './lvis_tfrecords/test*'
model_dir = './trained_model/'
export_dir ='./exported_model/'
if not os.path.exists(model_dir):
os.mkdir(model_dir)
In Model Garden, the collections of parameters that define a model are called configs. Model Garden can create a config based on a known set of parameters via a factory.
Use the retinanet_mobilenet_coco experiment configuration, as defined by tfm.vision.configs.maskrcnn.maskrcnn_mobilenet_coco.
Please find all the registered experiments here
The configuration defines an experiment to train a Mask R-CNN model with mobilenet as backbone and FPN as decoder. Default Configuration is trained on COCO train2017 and evaluated on COCO val2017.
There are also other alternative experiments available such as
maskrcnn_resnetfpn_coco,
maskrcnn_spinenet_coco and more. One can switch to them by changing the experiment name argument to the get_exp_config function.
exp_config = exp_factory.get_exp_config('maskrcnn_mobilenet_coco')
model_ckpt_path = './model_ckpt/'
if not os.path.exists(model_ckpt_path):
os.mkdir(model_ckpt_path)
!gsutil cp gs://tf_model_garden/vision/mobilenet/v2_1.0_float/ckpt-180648.data-00000-of-00001 './model_ckpt/'
!gsutil cp gs://tf_model_garden/vision/mobilenet/v2_1.0_float/ckpt-180648.index './model_ckpt/'
Copying gs://tf_model_garden/vision/mobilenet/v2_1.0_float/ckpt-180648.data-00000-of-00001... / [0 files][ 0.0 B/ 26.9 MiB] - - [1 files][ 26.9 MiB/ 26.9 MiB] Operation completed over 1 objects/26.9 MiB. Copying gs://tf_model_garden/vision/mobilenet/v2_1.0_float/ckpt-180648.index... / [0 files][ 0.0 B/ 7.5 KiB] / [1 files][ 7.5 KiB/ 7.5 KiB] Operation completed over 1 objects/7.5 KiB.
Adjust the model and dataset configurations so that it works with custom dataset.
BATCH_SIZE = 8
HEIGHT, WIDTH = 256, 256
IMG_SHAPE = [HEIGHT, WIDTH, 3]
# Backbone Config
exp_config.task.annotation_file = None
exp_config.task.freeze_backbone = True
exp_config.task.init_checkpoint = "./model_ckpt/ckpt-180648"
exp_config.task.init_checkpoint_modules = "backbone"
# Model Config
exp_config.task.model.num_classes = NUM_CLASSES + 1
exp_config.task.model.input_size = IMG_SHAPE
# Training Data Config
exp_config.task.train_data.input_path = train_data_input_path
exp_config.task.train_data.dtype = 'float32'
exp_config.task.train_data.global_batch_size = BATCH_SIZE
exp_config.task.train_data.shuffle_buffer_size = 64
exp_config.task.train_data.parser.aug_scale_max = 1.0
exp_config.task.train_data.parser.aug_scale_min = 1.0
# Validation Data Config
exp_config.task.validation_data.input_path = valid_data_input_path
exp_config.task.validation_data.dtype = 'float32'
exp_config.task.validation_data.global_batch_size = BATCH_SIZE
Adjust the trainer configuration.
logical_device_names = [logical_device.name for logical_device in tf.config.list_logical_devices()]
if 'GPU' in ''.join(logical_device_names):
print('This may be broken in Colab.')
device = 'GPU'
elif 'TPU' in ''.join(logical_device_names):
print('This may be broken in Colab.')
device = 'TPU'
else:
print('Running on CPU is slow, so only train for a few steps.')
device = 'CPU'
train_steps = 2000
exp_config.trainer.steps_per_loop = 200 # steps_per_loop = num_of_training_examples // train_batch_size
exp_config.trainer.summary_interval = 200
exp_config.trainer.checkpoint_interval = 200
exp_config.trainer.validation_interval = 200
exp_config.trainer.validation_steps = 200 # validation_steps = num_of_validation_examples // eval_batch_size
exp_config.trainer.train_steps = train_steps
exp_config.trainer.optimizer_config.warmup.linear.warmup_steps = 200
exp_config.trainer.optimizer_config.learning_rate.type = 'cosine'
exp_config.trainer.optimizer_config.learning_rate.cosine.decay_steps = train_steps
exp_config.trainer.optimizer_config.learning_rate.cosine.initial_learning_rate = 0.07
exp_config.trainer.optimizer_config.warmup.linear.warmup_learning_rate = 0.05
This may be broken in Colab.
Print the modified configuration.
pp.pprint(exp_config.as_dict())
display.Javascript("google.colab.output.setIframeHeight('500px');")
{ 'runtime': { 'all_reduce_alg': None,
'batchnorm_spatial_persistent': False,
'dataset_num_private_threads': None,
'default_shard_dim': -1,
'distribution_strategy': 'mirrored',
'enable_xla': False,
'gpu_thread_mode': None,
'loss_scale': None,
'mixed_precision_dtype': 'bfloat16',
'num_cores_per_replica': 1,
'num_gpus': 0,
'num_packs': 1,
'per_gpu_thread_count': 0,
'run_eagerly': False,
'task_index': -1,
'tpu': None,
'tpu_enable_xla_dynamic_padder': None,
'use_tpu_mp_strategy': False,
'worker_hosts': None},
'task': { 'allow_image_summary': False,
'allowed_mask_class_ids': None,
'annotation_file': None,
'differential_privacy_config': None,
'freeze_backbone': True,
'init_checkpoint': './model_ckpt/ckpt-180648',
'init_checkpoint_modules': 'backbone',
'losses': { 'class_weights': None,
'frcnn_box_weight': 1.0,
'frcnn_class_loss_top_k_percent': 1.0,
'frcnn_class_use_binary_cross_entropy': False,
'frcnn_class_weight': 1.0,
'frcnn_huber_loss_delta': 1.0,
'l2_weight_decay': 4e-05,
'loss_weight': 1.0,
'mask_weight': 1.0,
'rpn_box_weight': 1.0,
'rpn_huber_loss_delta': 0.1111111111111111,
'rpn_score_weight': 1.0},
'model': { 'anchor': { 'anchor_size': 3,
'aspect_ratios': [0.5, 1.0, 2.0],
'num_scales': 1},
'backbone': { 'mobilenet': { 'filter_size_scale': 1.0,
'flat_stochastic_depth_drop_rate': True,
'model_id': 'MobileNetV2',
'output_intermediate_endpoints': False,
'output_stride': None,
'stochastic_depth_drop_rate': 0.0},
'type': 'mobilenet'},
'decoder': { 'fpn': { 'fusion_type': 'sum',
'num_filters': 128,
'use_keras_layer': False,
'use_separable_conv': True},
'type': 'fpn'},
'detection_generator': { 'apply_nms': True,
'max_num_detections': 100,
'nms_iou_threshold': 0.5,
'nms_version': 'v2',
'pre_nms_score_threshold': 0.05,
'pre_nms_top_k': 5000,
'soft_nms_sigma': None,
'use_cpu_nms': False,
'use_sigmoid_probability': False},
'detection_head': { 'cascade_class_ensemble': False,
'class_agnostic_bbox_pred': False,
'fc_dims': 512,
'num_convs': 4,
'num_fcs': 1,
'num_filters': 128,
'use_separable_conv': True},
'include_mask': True,
'input_size': [256, 256, 3],
'mask_head': { 'class_agnostic': False,
'num_convs': 4,
'num_filters': 128,
'upsample_factor': 2,
'use_separable_conv': True},
'mask_roi_aligner': { 'crop_size': 14,
'sample_offset': 0.5},
'mask_sampler': {'num_sampled_masks': 128},
'max_level': 6,
'min_level': 3,
'norm_activation': { 'activation': 'relu6',
'norm_epsilon': 0.001,
'norm_momentum': 0.99,
'use_sync_bn': True},
'num_classes': 4,
'outer_boxes_scale': 1.0,
'roi_aligner': { 'crop_size': 7,
'sample_offset': 0.5},
'roi_generator': { 'nms_iou_threshold': 0.7,
'num_proposals': 1000,
'pre_nms_min_size_threshold': 0.0,
'pre_nms_score_threshold': 0.0,
'pre_nms_top_k': 2000,
'test_nms_iou_threshold': 0.7,
'test_num_proposals': 1000,
'test_pre_nms_min_size_threshold': 0.0,
'test_pre_nms_score_threshold': 0.0,
'test_pre_nms_top_k': 1000,
'use_batched_nms': False},
'roi_sampler': { 'background_iou_high_threshold': 0.5,
'background_iou_low_threshold': 0.0,
'cascade_iou_thresholds': None,
'foreground_fraction': 0.25,
'foreground_iou_threshold': 0.5,
'mix_gt_boxes': True,
'num_sampled_rois': 512},
'rpn_head': { 'num_convs': 1,
'num_filters': 128,
'use_separable_conv': True} },
'name': None,
'per_category_metrics': False,
'train_data': { 'apply_tf_data_service_before_batching': False,
'autotune_algorithm': None,
'block_length': 1,
'cache': False,
'cycle_length': None,
'decoder': { 'simple_decoder': { 'attribute_names': [ ],
'mask_binarize_threshold': None,
'regenerate_source_id': False},
'type': 'simple_decoder'},
'deterministic': None,
'drop_remainder': True,
'dtype': 'float32',
'enable_shared_tf_data_service_between_parallel_trainers': False,
'enable_tf_data_service': False,
'file_type': 'tfrecord',
'global_batch_size': 8,
'input_path': './lvis_tfrecords/train*',
'is_training': True,
'num_examples': -1,
'parser': { 'aug_rand_hflip': True,
'aug_rand_vflip': False,
'aug_scale_max': 1.0,
'aug_scale_min': 1.0,
'aug_type': None,
'keep_aspect_ratio': True,
'mask_crop_size': 112,
'match_threshold': 0.5,
'max_num_instances': 100,
'num_channels': 3,
'pad': True,
'rpn_batch_size_per_im': 256,
'rpn_fg_fraction': 0.5,
'rpn_match_threshold': 0.7,
'rpn_unmatched_threshold': 0.3,
'skip_crowd_during_training': True,
'unmatched_threshold': 0.5},
'prefetch_buffer_size': None,
'ram_budget': None,
'seed': None,
'sharding': True,
'shuffle_buffer_size': 64,
'tf_data_service_address': None,
'tf_data_service_job_name': None,
'tfds_as_supervised': False,
'tfds_data_dir': '',
'tfds_name': '',
'tfds_skip_decoding_feature': '',
'tfds_split': '',
'trainer_id': None,
'weights': None},
'use_approx_instance_metrics': False,
'use_coco_metrics': True,
'use_wod_metrics': False,
'validation_data': { 'apply_tf_data_service_before_batching': False,
'autotune_algorithm': None,
'block_length': 1,
'cache': False,
'cycle_length': None,
'decoder': { 'simple_decoder': { 'attribute_names': [ ],
'mask_binarize_threshold': None,
'regenerate_source_id': False},
'type': 'simple_decoder'},
'deterministic': None,
'drop_remainder': False,
'dtype': 'float32',
'enable_shared_tf_data_service_between_parallel_trainers': False,
'enable_tf_data_service': False,
'file_type': 'tfrecord',
'global_batch_size': 8,
'input_path': './lvis_tfrecords/val*',
'is_training': False,
'num_examples': -1,
'parser': { 'aug_rand_hflip': False,
'aug_rand_vflip': False,
'aug_scale_max': 1.0,
'aug_scale_min': 1.0,
'aug_type': None,
'keep_aspect_ratio': True,
'mask_crop_size': 112,
'match_threshold': 0.5,
'max_num_instances': 100,
'num_channels': 3,
'pad': True,
'rpn_batch_size_per_im': 256,
'rpn_fg_fraction': 0.5,
'rpn_match_threshold': 0.7,
'rpn_unmatched_threshold': 0.3,
'skip_crowd_during_training': True,
'unmatched_threshold': 0.5},
'prefetch_buffer_size': None,
'ram_budget': None,
'seed': None,
'sharding': True,
'shuffle_buffer_size': 10000,
'tf_data_service_address': None,
'tf_data_service_job_name': None,
'tfds_as_supervised': False,
'tfds_data_dir': '',
'tfds_name': '',
'tfds_skip_decoding_feature': '',
'tfds_split': '',
'trainer_id': None,
'weights': None} },
'trainer': { 'allow_tpu_summary': False,
'best_checkpoint_eval_metric': '',
'best_checkpoint_export_subdir': '',
'best_checkpoint_metric_comp': 'higher',
'checkpoint_interval': 200,
'continuous_eval_timeout': 3600,
'eval_tf_function': True,
'eval_tf_while_loop': False,
'loss_upper_bound': 1000000.0,
'max_to_keep': 5,
'optimizer_config': { 'ema': None,
'learning_rate': { 'cosine': { 'alpha': 0.0,
'decay_steps': 2000,
'initial_learning_rate': 0.07,
'name': 'CosineDecay',
'offset': 0},
'type': 'cosine'},
'optimizer': { 'sgd': { 'clipnorm': None,
'clipvalue': None,
'decay': 0.0,
'global_clipnorm': None,
'momentum': 0.9,
'name': 'SGD',
'nesterov': False},
'type': 'sgd'},
'warmup': { 'linear': { 'name': 'linear',
'warmup_learning_rate': 0.05,
'warmup_steps': 200},
'type': 'linear'} },
'preemption_on_demand_checkpoint': True,
'recovery_begin_steps': 0,
'recovery_max_trials': 0,
'steps_per_loop': 200,
'summary_interval': 200,
'train_steps': 2000,
'train_tf_function': True,
'train_tf_while_loop': True,
'validation_interval': 200,
'validation_steps': 200,
'validation_summary_subdir': 'validation'} }
<IPython.core.display.Javascript object>
Set up the distribution strategy.
# Setting up the Strategy
if exp_config.runtime.mixed_precision_dtype == tf.float16:
tf.keras.mixed_precision.set_global_policy('mixed_float16')
if 'GPU' in ''.join(logical_device_names):
distribution_strategy = tf.distribute.MirroredStrategy()
elif 'TPU' in ''.join(logical_device_names):
tf.tpu.experimental.initialize_tpu_system()
tpu = tf.distribute.cluster_resolver.TPUClusterResolver(tpu='/device:TPU_SYSTEM:0')
distribution_strategy = tf.distribute.experimental.TPUStrategy(tpu)
else:
print('Warning: this will be really slow.')
distribution_strategy = tf.distribute.OneDeviceStrategy(logical_device_names[0])
print("Done")
INFO:tensorflow:Using MirroredStrategy with devices ('/job:localhost/replica:0/task:0/device:GPU:0', '/job:localhost/replica:0/task:0/device:GPU:1', '/job:localhost/replica:0/task:0/device:GPU:2', '/job:localhost/replica:0/task:0/device:GPU:3')
Done
Create the Task object (tfm.core.base_task.Task) from the config_definitions.TaskConfig.
The Task object has all the methods necessary for building the dataset, building the model, and running training & evaluation. These methods are driven by tfm.core.train_lib.run_experiment.
with distribution_strategy.scope():
task = tfm.core.task_factory.get_task(exp_config.task, logging_dir=model_dir)
Visualize a batch of the data.
for images, labels in task.build_inputs(exp_config.task.train_data).take(1):
print()
print(f'images.shape: {str(images.shape):16} images.dtype: {images.dtype!r}')
print(f'labels.keys: {labels.keys()}')
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.11/site-packages/tensorflow/python/util/deprecation.py:660: calling map_fn_v2 (from tensorflow.python.ops.map_fn) with dtype is deprecated and will be removed in a future version.
Instructions for updating:
Use fn_output_signature instead
W0000 00:00:1793448003.228166 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -56 } dim { size: -42 } dim { size: -43 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -3 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -3 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -3 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
images.shape: (8, 256, 256, 3) images.dtype: tf.float32
labels.keys: dict_keys(['anchor_boxes', 'image_info', 'rpn_score_targets', 'rpn_box_targets', 'gt_boxes', 'gt_classes', 'gt_outer_boxes', 'gt_masks'])
Create Category Index Dictionary to map the labels to corresponding label names
tf_ex_decoder = TfExampleDecoder(include_mask=True)
Helper Function for Visualizing the results from TFRecords
Use visualize_boxes_and_labels_on_image_array from visualization_utils to draw bounding boxes on the image.
def show_batch(raw_records):
plt.figure(figsize=(20, 20))
use_normalized_coordinates=True
min_score_thresh = 0.30
for i, serialized_example in enumerate(raw_records):
plt.subplot(1, 3, i + 1)
decoded_tensors = tf_ex_decoder.decode(serialized_example)
image = decoded_tensors['image'].numpy().astype('uint8')
scores = np.ones(shape=(len(decoded_tensors['groundtruth_boxes'])))
# print(decoded_tensors['groundtruth_instance_masks'].numpy().shape)
# print(decoded_tensors.keys())
visualization_utils.visualize_boxes_and_labels_on_image_array(
image,
decoded_tensors['groundtruth_boxes'].numpy(),
decoded_tensors['groundtruth_classes'].numpy().astype('int'),
scores,
category_index=category_index,
use_normalized_coordinates=use_normalized_coordinates,
min_score_thresh=min_score_thresh,
instance_masks=decoded_tensors['groundtruth_instance_masks'].numpy().astype('uint8'),
line_thickness=4)
plt.imshow(image)
plt.axis("off")
plt.title(f"Image-{i+1}")
plt.show()
Visualization of Train Data
The bounding box detection has three components
- Class label of the object detected.
- Percentage of match between predicted and ground truth bounding boxes.
- Instance Segmentation Mask
buffer_size = 100
num_of_examples = 3
train_tfrecords = tf.io.gfile.glob(exp_config.task.train_data.input_path)
raw_records = tf.data.TFRecordDataset(train_tfrecords).shuffle(buffer_size=buffer_size).take(num_of_examples)
show_batch(raw_records)
Train and evaluate
We follow the COCO challenge tradition to evaluate the accuracy of object detection based on mAP(mean Average Precision). Please check here for detail explanation of how evaluation metrics for detection task is done.
IoU: is defined as the area of the intersection divided by the area of the union of a predicted bounding box and ground truth bounding box.
model, eval_logs = tfm.core.train_lib.run_experiment(
distribution_strategy=distribution_strategy,
task=task,
mode='train_and_eval',
params=exp_config,
model_dir=model_dir,
run_post_eval=True)
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
INFO:tensorflow:Reduce to /job:localhost/replica:0/task:0/device:CPU:0 then broadcast to ('/job:localhost/replica:0/task:0/device:CPU:0',).
WARNING:tensorflow:`tf.keras.layers.experimental.SyncBatchNormalization` endpoint is deprecated and will be removed in a future release. Please use `tf.keras.layers.BatchNormalization` with parameter `synchronized` set to True.
/tmpfs/src/tf_docs_env/lib/python3.11/site-packages/tf_keras/src/engine/functional.py:641: UserWarning: Input dict contained keys ['6'] which did not match any model input. They will be ignored by the model.
inputs = self._flatten_to_reference_inputs(inputs)
WARNING:tensorflow:`tf.keras.layers.experimental.SyncBatchNormalization` endpoint is deprecated and will be removed in a future release. Please use `tf.keras.layers.BatchNormalization` with parameter `synchronized` set to True.
/tmpfs/src/tf_docs_env/lib/python3.11/site-packages/tf_keras/src/initializers/initializers.py:121: UserWarning: The initializer VarianceScaling is unseeded and being called multiple times, which will return identical values each time (even if the initializer is unseeded). Please update your code to provide a seed to the initializer, or avoid using the same initializer instance more than once.
warnings.warn(
loading annotations into memory...
Done (t=0.01s)
creating index...
index created!
restoring or initializing model...
INFO:tensorflow:Customized initialization is done through the passed `init_fn`.
INFO:tensorflow:Customized initialization is done through the passed `init_fn`.
train | step: 0 | training until step 200...
W0000 00:00:1793448031.844539 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -56 } dim { size: -42 } dim { size: -43 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -3 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -3 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -3 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 101 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 101 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 1 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 101 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
INFO:tensorflow:Collective all_reduce tensors: 101 all_reduces, num_devices = 4, group_size = 4, implementation = CommunicationImplementation.NCCL, num_packs = 1
train | step: 200 | steps/sec: 1.3 | output:
{'frcnn_box_loss': np.float32(0.3028656),
'frcnn_cls_loss': np.float32(0.05610391),
'learning_rate': np.float32(0.06828698),
'mask_loss': np.float32(0.5498602),
'model_loss': np.float32(1.0419443),
'rpn_box_loss': np.float32(0.058857378),
'rpn_score_loss': np.float32(0.07425722),
'total_loss': np.float32(1.3438727),
'training_loss': np.float32(1.3438727)}
saved checkpoint to ./trained_model/ckpt-200.
eval | step: 200 | running 200 steps of evaluation...
W0000 00:00:1793448183.644006 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=1.19s).
Accumulating evaluation results...
DONE (t=0.30s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.004
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.013
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.002
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.001
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.015
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.008
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.037
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.057
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.003
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.054
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.134
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=1.36s).
Accumulating evaluation results...
DONE (t=0.30s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.003
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.012
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.013
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.008
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.026
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.036
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.019
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.104
eval | step: 200 | steps/sec: 2.6 | eval time: 76.8 sec | output:
{'AP': np.float32(0.0037186528),
'AP50': np.float32(0.013320773),
'AP75': np.float32(0.0018170512),
'APl': np.float32(0.015024962),
'APm': np.float32(0.0013786417),
'APs': np.float32(6.6047694e-05),
'ARl': np.float32(0.13370787),
'ARm': np.float32(0.054248367),
'ARmax1': np.float32(0.008400646),
'ARmax10': np.float32(0.03654923),
'ARmax100': np.float32(0.05690464),
'ARs': np.float32(0.0033783785),
'mask_AP': np.float32(0.0025623788),
'mask_AP50': np.float32(0.011915229),
'mask_AP75': np.float32(2.4009856e-05),
'mask_APl': np.float32(0.012821915),
'mask_APm': np.float32(0.00040607186),
'mask_APs': np.float32(3.4378438e-06),
'mask_ARl': np.float32(0.10374532),
'mask_ARm': np.float32(0.019444445),
'mask_ARmax1': np.float32(0.007969844),
'mask_ARmax10': np.float32(0.025563769),
'mask_ARmax100': np.float32(0.036064576),
'mask_ARs': np.float32(0.00022522523),
'steps_per_second': 2.603593683369802,
'validation_loss': np.float32(0.0)}
train | step: 200 | training until step 400...
train | step: 400 | steps/sec: 1.8 | output:
{'frcnn_box_loss': np.float32(0.3115651),
'frcnn_cls_loss': np.float32(0.04525192),
'learning_rate': np.float32(0.06331559),
'mask_loss': np.float32(0.42246544),
'model_loss': np.float32(0.8524089),
'rpn_box_loss': np.float32(0.053396776),
'rpn_score_loss': np.float32(0.01972979),
'total_loss': np.float32(1.1536005),
'training_loss': np.float32(1.1536005)}
saved checkpoint to ./trained_model/ckpt-400.
eval | step: 400 | running 200 steps of evaluation...
W0000 00:00:1793448296.496547 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.68s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.033
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.101
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.012
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.015
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.095
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.041
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.070
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.074
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.003
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.051
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.194
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=0.76s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.018
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.067
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.002
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.005
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.058
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.026
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.038
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.039
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.013
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.121
eval | step: 400 | steps/sec: 4.5 | eval time: 44.3 sec | output:
{'AP': np.float32(0.03288171),
'AP50': np.float32(0.10108758),
'AP75': np.float32(0.012047551),
'APl': np.float32(0.09541191),
'APm': np.float32(0.015482943),
'APs': np.float32(0.00031868793),
'ARl': np.float32(0.19400749),
'ARm': np.float32(0.05147059),
'ARmax1': np.float32(0.041087776),
'ARmax10': np.float32(0.070382334),
'ARmax100': np.float32(0.07355951),
'ARs': np.float32(0.0033783785),
'mask_AP': np.float32(0.017617872),
'mask_AP50': np.float32(0.0667538),
'mask_AP75': np.float32(0.002162287),
'mask_APl': np.float32(0.05771005),
'mask_APm': np.float32(0.005087531),
'mask_APs': np.float32(3.0278256e-06),
'mask_ARl': np.float32(0.121348314),
'mask_ARm': np.float32(0.013235294),
'mask_ARmax1': np.float32(0.025955843),
'mask_ARmax10': np.float32(0.03839526),
'mask_ARmax100': np.float32(0.039310716),
'mask_ARs': np.float32(0.00022522523),
'steps_per_second': 4.5108866159461005,
'validation_loss': np.float32(0.0)}
train | step: 400 | training until step 600...
train | step: 600 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.27542165),
'frcnn_cls_loss': np.float32(0.04818673),
'learning_rate': np.float32(0.055572484),
'mask_loss': np.float32(0.38600853),
'model_loss': np.float32(0.7809595),
'rpn_box_loss': np.float32(0.052467704),
'rpn_score_loss': np.float32(0.018874882),
'total_loss': np.float32(1.0811303),
'training_loss': np.float32(1.0811303)}
saved checkpoint to ./trained_model/ckpt-600.
eval | step: 600 | running 200 steps of evaluation...
W0000 00:00:1793448376.560873 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.79s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.061
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.147
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.035
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.003
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.025
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.180
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.067
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.093
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.100
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.018
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.074
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.250
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=0.91s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.025
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.093
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.004
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.009
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.077
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.034
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.043
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.044
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.001
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.022
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.127
eval | step: 600 | steps/sec: 4.6 | eval time: 43.5 sec | output:
{'AP': np.float32(0.06149739),
'AP50': np.float32(0.14658919),
'AP75': np.float32(0.034732983),
'APl': np.float32(0.17982994),
'APm': np.float32(0.024727613),
'APs': np.float32(0.0025991478),
'ARl': np.float32(0.24981274),
'ARm': np.float32(0.07401961),
'ARmax1': np.float32(0.066720515),
'ARmax10': np.float32(0.09337641),
'ARmax100': np.float32(0.10037695),
'ARs': np.float32(0.017792793),
'mask_AP': np.float32(0.025269493),
'mask_AP50': np.float32(0.09316484),
'mask_AP75': np.float32(0.004277595),
'mask_APl': np.float32(0.07722846),
'mask_APm': np.float32(0.009491281),
'mask_APs': np.float32(9.690565e-06),
'mask_ARl': np.float32(0.1269663),
'mask_ARm': np.float32(0.021895425),
'mask_ARmax1': np.float32(0.033817984),
'mask_ARmax10': np.float32(0.042595584),
'mask_ARmax100': np.float32(0.043995693),
'mask_ARs': np.float32(0.0011261262),
'steps_per_second': 4.593099320276593,
'validation_loss': np.float32(0.0)}
train | step: 600 | training until step 800...
train | step: 800 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.276611),
'frcnn_cls_loss': np.float32(0.043674573),
'learning_rate': np.float32(0.045815594),
'mask_loss': np.float32(0.38377562),
'model_loss': np.float32(0.7682332),
'rpn_box_loss': np.float32(0.04824703),
'rpn_score_loss': np.float32(0.01592489),
'total_loss': np.float32(1.0674156),
'training_loss': np.float32(1.0674156)}
saved checkpoint to ./trained_model/ckpt-800.
eval | step: 800 | running 200 steps of evaluation...
W0000 00:00:1793448455.789217 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.66s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.068
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.147
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.051
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.004
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.038
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.190
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.075
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.099
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.104
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.017
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.072
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.264
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=0.74s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.044
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.122
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.017
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.018
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.133
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.053
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.064
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.065
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.001
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.029
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.191
eval | step: 800 | steps/sec: 4.5 | eval time: 44.3 sec | output:
{'AP': np.float32(0.06764329),
'AP50': np.float32(0.1465125),
'AP75': np.float32(0.050791476),
'APl': np.float32(0.18984962),
'APm': np.float32(0.037950225),
'APs': np.float32(0.0041238493),
'ARl': np.float32(0.26404494),
'ARm': np.float32(0.072222225),
'ARmax1': np.float32(0.07522886),
'ARmax10': np.float32(0.099461496),
'ARmax100': np.float32(0.10360797),
'ARs': np.float32(0.016891891),
'mask_AP': np.float32(0.04436627),
'mask_AP50': np.float32(0.12203395),
'mask_AP75': np.float32(0.017342584),
'mask_APl': np.float32(0.13282272),
'mask_APm': np.float32(0.018245617),
'mask_APs': np.float32(0.00023044994),
'mask_ARl': np.float32(0.1911985),
'mask_ARm': np.float32(0.02875817),
'mask_ARmax1': np.float32(0.05250404),
'mask_ARmax10': np.float32(0.064028),
'mask_ARmax100': np.float32(0.064781904),
'mask_ARs': np.float32(0.0013513514),
'steps_per_second': 4.518995376882406,
'validation_loss': np.float32(0.0)}
train | step: 800 | training until step 1000...
train | step: 1000 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.25857982),
'frcnn_cls_loss': np.float32(0.043173894),
'learning_rate': np.float32(0.034999996),
'mask_loss': np.float32(0.3763401),
'model_loss': np.float32(0.7385902),
'rpn_box_loss': np.float32(0.044937775),
'rpn_score_loss': np.float32(0.015558512),
'total_loss': np.float32(1.0369316),
'training_loss': np.float32(1.0369316)}
saved checkpoint to ./trained_model/ckpt-1000.
eval | step: 1000 | running 200 steps of evaluation...
W0000 00:00:1793448535.738566 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.70s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.080
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.163
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.066
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.007
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.045
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.219
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.083
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.109
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.115
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.022
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.092
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.276
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=0.80s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.051
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.132
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.023
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.019
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.152
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.057
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.068
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.069
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.002
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.036
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.199
eval | step: 1000 | steps/sec: 4.6 | eval time: 43.0 sec | output:
{'AP': np.float32(0.08030654),
'AP50': np.float32(0.16319275),
'AP75': np.float32(0.0655098),
'APl': np.float32(0.21911536),
'APm': np.float32(0.04531654),
'APs': np.float32(0.0068145213),
'ARl': np.float32(0.2758427),
'ARm': np.float32(0.09166667),
'ARmax1': np.float32(0.082929455),
'ARmax10': np.float32(0.1091007),
'ARmax100': np.float32(0.11470113),
'ARs': np.float32(0.021846848),
'mask_AP': np.float32(0.050672267),
'mask_AP50': np.float32(0.13234015),
'mask_AP75': np.float32(0.02292493),
'mask_APl': np.float32(0.15201184),
'mask_APm': np.float32(0.019453527),
'mask_APs': np.float32(0.00018186902),
'mask_ARl': np.float32(0.19906367),
'mask_ARm': np.float32(0.035620917),
'mask_ARmax1': np.float32(0.056758214),
'mask_ARmax10': np.float32(0.068174474),
'mask_ARmax100': np.float32(0.06941303),
'mask_ARs': np.float32(0.0018018018),
'steps_per_second': 4.647639780392221,
'validation_loss': np.float32(0.0)}
train | step: 1000 | training until step 1200...
train | step: 1200 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.24661812),
'frcnn_cls_loss': np.float32(0.04405328),
'learning_rate': np.float32(0.024184398),
'mask_loss': np.float32(0.36619195),
'model_loss': np.float32(0.7185364),
'rpn_box_loss': np.float32(0.046044096),
'rpn_score_loss': np.float32(0.015628984),
'total_loss': np.float32(1.0162216),
'training_loss': np.float32(1.0162216)}
saved checkpoint to ./trained_model/ckpt-1200.
eval | step: 1200 | running 200 steps of evaluation...
W0000 00:00:1793448614.551699 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.90s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.083
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.168
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.074
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.008
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.043
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.227
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.084
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.112
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.121
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.034
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.096
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.284
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=1.03s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.048
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.127
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.019
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.016
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.146
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.055
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.064
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.066
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.003
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.031
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.190
eval | step: 1200 | steps/sec: 4.4 | eval time: 45.4 sec | output:
{'AP': np.float32(0.08303887),
'AP50': np.float32(0.16841075),
'AP75': np.float32(0.07427567),
'APl': np.float32(0.22686976),
'APm': np.float32(0.042609617),
'APs': np.float32(0.0075140847),
'ARl': np.float32(0.28408238),
'ARm': np.float32(0.095915034),
'ARmax1': np.float32(0.08449111),
'ARmax10': np.float32(0.11200862),
'ARmax100': np.float32(0.12132472),
'ARs': np.float32(0.03355856),
'mask_AP': np.float32(0.047515236),
'mask_AP50': np.float32(0.12688565),
'mask_AP75': np.float32(0.019116431),
'mask_APl': np.float32(0.14603789),
'mask_APm': np.float32(0.016171204),
'mask_APs': np.float32(0.00016250614),
'mask_ARl': np.float32(0.19026217),
'mask_ARm': np.float32(0.030882353),
'mask_ARmax1': np.float32(0.05508885),
'mask_ARmax10': np.float32(0.06386645),
'mask_ARmax100': np.float32(0.06553581),
'mask_ARs': np.float32(0.0027027028),
'steps_per_second': 4.402650820903657,
'validation_loss': np.float32(0.0)}
train | step: 1200 | training until step 1400...
train | step: 1400 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.24072166),
'frcnn_cls_loss': np.float32(0.04367125),
'learning_rate': np.float32(0.014427517),
'mask_loss': np.float32(0.35957423),
'model_loss': np.float32(0.7013161),
'rpn_box_loss': np.float32(0.04223218),
'rpn_score_loss': np.float32(0.01511682),
'total_loss': np.float32(0.9985277),
'training_loss': np.float32(0.9985277)}
saved checkpoint to ./trained_model/ckpt-1400.
eval | step: 1400 | running 200 steps of evaluation...
W0000 00:00:1793448695.730132 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.92s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.088
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.174
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.080
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.007
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.050
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.237
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.088
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.117
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.126
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.033
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.104
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.291
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=1.01s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.051
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.133
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.027
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.019
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.154
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.059
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.069
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.071
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.004
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.037
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.203
eval | step: 1400 | steps/sec: 4.5 | eval time: 44.7 sec | output:
{'AP': np.float32(0.088213295),
'AP50': np.float32(0.17405045),
'AP75': np.float32(0.08049665),
'APl': np.float32(0.23672514),
'APm': np.float32(0.049936596),
'APs': np.float32(0.006778566),
'ARl': np.float32(0.29082397),
'ARm': np.float32(0.104411766),
'ARmax1': np.float32(0.087937534),
'ARmax10': np.float32(0.117016695),
'ARmax100': np.float32(0.12595585),
'ARs': np.float32(0.033108108),
'mask_AP': np.float32(0.05096652),
'mask_AP50': np.float32(0.13347904),
'mask_AP75': np.float32(0.027180659),
'mask_APl': np.float32(0.15414236),
'mask_APm': np.float32(0.019091649),
'mask_APs': np.float32(0.00038800674),
'mask_ARl': np.float32(0.20280899),
'mask_ARm': np.float32(0.036928106),
'mask_ARmax1': np.float32(0.058589123),
'mask_ARmax10': np.float32(0.068605274),
'mask_ARmax100': np.float32(0.07140549),
'mask_ARs': np.float32(0.003828829),
'steps_per_second': 4.47492476437892,
'validation_loss': np.float32(0.0)}
train | step: 1400 | training until step 1600...
train | step: 1600 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.24547271),
'frcnn_cls_loss': np.float32(0.04113173),
'learning_rate': np.float32(0.006684403),
'mask_loss': np.float32(0.3597195),
'model_loss': np.float32(0.6987518),
'rpn_box_loss': np.float32(0.038811594),
'rpn_score_loss': np.float32(0.013616253),
'total_loss': np.float32(0.9956663),
'training_loss': np.float32(0.9956663)}
saved checkpoint to ./trained_model/ckpt-1600.
eval | step: 1600 | running 200 steps of evaluation...
W0000 00:00:1793448776.122117 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.66s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.088
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.170
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.086
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.008
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.051
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.239
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.089
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.117
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.123
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.029
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.096
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.295
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=0.74s).
Accumulating evaluation results...
DONE (t=0.23s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.056
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.139
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.033
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.022
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.168
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.061
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.072
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.073
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.003
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.035
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.212
eval | step: 1600 | steps/sec: 4.7 | eval time: 42.8 sec | output:
{'AP': np.float32(0.08806156),
'AP50': np.float32(0.17035413),
'AP75': np.float32(0.08632451),
'APl': np.float32(0.23880145),
'APm': np.float32(0.05121247),
'APs': np.float32(0.0076173404),
'ARl': np.float32(0.2949438),
'ARm': np.float32(0.095751636),
'ARmax1': np.float32(0.08858374),
'ARmax10': np.float32(0.117447495),
'ARmax100': np.float32(0.12320948),
'ARs': np.float32(0.02882883),
'mask_AP': np.float32(0.056023046),
'mask_AP50': np.float32(0.13898644),
'mask_AP75': np.float32(0.033470217),
'mask_APl': np.float32(0.16831647),
'mask_APm': np.float32(0.021727486),
'mask_APs': np.float32(0.0003525649),
'mask_ARl': np.float32(0.21198502),
'mask_ARm': np.float32(0.035457518),
'mask_ARmax1': np.float32(0.060904685),
'mask_ARmax10': np.float32(0.07199784),
'mask_ARmax100': np.float32(0.07329025),
'mask_ARs': np.float32(0.0027027028),
'steps_per_second': 4.675441714311834,
'validation_loss': np.float32(0.0)}
train | step: 1600 | training until step 1800...
train | step: 1800 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.23522289),
'frcnn_cls_loss': np.float32(0.042658806),
'learning_rate': np.float32(0.0017130232),
'mask_loss': np.float32(0.36148757),
'model_loss': np.float32(0.6939177),
'rpn_box_loss': np.float32(0.040340208),
'rpn_score_loss': np.float32(0.014208256),
'total_loss': np.float32(0.9906848),
'training_loss': np.float32(0.9906848)}
saved checkpoint to ./trained_model/ckpt-1800.
eval | step: 1800 | running 200 steps of evaluation...
W0000 00:00:1793448854.649672 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.82s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.093
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.180
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.090
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.009
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.055
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.247
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.091
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.120
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.128
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.033
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.106
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.297
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=0.93s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.055
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.137
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.028
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.021
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.165
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.060
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.069
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.071
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.004
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.037
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.203
eval | step: 1800 | steps/sec: 4.5 | eval time: 44.7 sec | output:
{'AP': np.float32(0.092969716),
'AP50': np.float32(0.18009107),
'AP75': np.float32(0.090151876),
'APl': np.float32(0.24698749),
'APm': np.float32(0.055247277),
'APs': np.float32(0.009420988),
'ARl': np.float32(0.29681647),
'ARm': np.float32(0.10637255),
'ARmax1': np.float32(0.09127625),
'ARmax10': np.float32(0.11992461),
'ARmax100': np.float32(0.12816371),
'ARs': np.float32(0.032657657),
'mask_AP': np.float32(0.05476431),
'mask_AP50': np.float32(0.13704266),
'mask_AP75': np.float32(0.02826284),
'mask_APl': np.float32(0.16541827),
'mask_APm': np.float32(0.020903315),
'mask_APs': np.float32(0.00047454107),
'mask_ARl': np.float32(0.20299625),
'mask_ARm': np.float32(0.036928106),
'mask_ARmax1': np.float32(0.05977383),
'mask_ARmax10': np.float32(0.06914378),
'mask_ARmax100': np.float32(0.07140549),
'mask_ARs': np.float32(0.0036036037),
'steps_per_second': 4.478597246977088,
'validation_loss': np.float32(0.0)}
train | step: 1800 | training until step 2000...
train | step: 2000 | steps/sec: 2.5 | output:
{'frcnn_box_loss': np.float32(0.22321194),
'frcnn_cls_loss': np.float32(0.04425861),
'learning_rate': np.float32(0.0),
'mask_loss': np.float32(0.34897017),
'model_loss': np.float32(0.6742164),
'rpn_box_loss': np.float32(0.04186353),
'rpn_score_loss': np.float32(0.01591219),
'total_loss': np.float32(0.97093797),
'training_loss': np.float32(0.97093797)}
saved checkpoint to ./trained_model/ckpt-2000.
eval | step: 2000 | running 200 steps of evaluation...
W0000 00:00:1793448935.113941 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.78s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.094
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.181
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.091
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.010
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.056
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.249
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.092
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.121
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.129
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.033
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.106
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.299
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=2.11s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.057
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.141
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.032
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.001
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.022
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.171
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.062
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.072
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.074
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.004
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.040
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.209
eval | step: 2000 | steps/sec: 4.5 | eval time: 44.0 sec | output:
{'AP': np.float32(0.09387283),
'AP50': np.float32(0.1813329),
'AP75': np.float32(0.090727),
'APl': np.float32(0.24916865),
'APm': np.float32(0.05625988),
'APs': np.float32(0.009651893),
'ARl': np.float32(0.2988764),
'ARm': np.float32(0.10637255),
'ARmax1': np.float32(0.09229941),
'ARmax10': np.float32(0.12094776),
'ARmax100': np.float32(0.12880991),
'ARs': np.float32(0.032882884),
'mask_AP': np.float32(0.05688256),
'mask_AP50': np.float32(0.14087473),
'mask_AP75': np.float32(0.03236862),
'mask_APl': np.float32(0.17055447),
'mask_APm': np.float32(0.022371016),
'mask_APs': np.float32(0.0006784558),
'mask_ARl': np.float32(0.2093633),
'mask_ARm': np.float32(0.039542485),
'mask_ARmax1': np.float32(0.06192784),
'mask_ARmax10': np.float32(0.071944),
'mask_ARmax100': np.float32(0.074098006),
'mask_ARs': np.float32(0.0036036037),
'steps_per_second': 4.547250561501104,
'validation_loss': np.float32(0.0)}
eval | step: 2000 | running 200 steps of evaluation...
W0000 00:00:1793448979.162871 41008 op_level_cost_estimator.cc:699] Error in PredictCost() for the op: op: "CropAndResize" attr { key: "T" value { type: DT_FLOAT } } attr { key: "extrapolation_value" value { f: 0 } } attr { key: "method" value { s: "bilinear" } } inputs { dtype: DT_FLOAT shape { dim { size: -5 } dim { size: -6 } dim { size: -7 } dim { size: 1 } } } inputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 4 } } } inputs { dtype: DT_INT32 shape { dim { size: -2 } } } inputs { dtype: DT_INT32 shape { dim { size: 2 } } value { dtype: DT_INT32 tensor_shape { dim { size: 2 } } int_val: 112 } } device { type: "CPU" vendor: "GenuineIntel" model: "111" frequency: 2299 num_cores: 32 environment { key: "cpu_instruction_set" value: "AVX SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2" } environment { key: "eigen" value: "3.4.90" } l1_cache_size: 32768 l2_cache_size: 262144 l3_cache_size: 47185920 memory_size: 268435456 } outputs { dtype: DT_FLOAT shape { dim { size: -2 } dim { size: 112 } dim { size: 112 } dim { size: 1 } } }
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.76s).
Accumulating evaluation results...
DONE (t=0.25s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.094
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.181
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.091
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.010
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.056
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.249
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.092
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.121
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.129
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.033
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.106
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.299
Running per image evaluation...
Evaluate annotation type *segm*
DONE (t=0.87s).
Accumulating evaluation results...
DONE (t=0.24s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.057
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.141
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.032
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.001
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.022
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.171
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.062
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.072
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.074
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.004
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.040
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.209
eval | step: 2000 | steps/sec: 4.6 | eval time: 43.5 sec | output:
{'AP': np.float32(0.09387283),
'AP50': np.float32(0.1813329),
'AP75': np.float32(0.090727),
'APl': np.float32(0.24916865),
'APm': np.float32(0.05625988),
'APs': np.float32(0.009651893),
'ARl': np.float32(0.2988764),
'ARm': np.float32(0.10637255),
'ARmax1': np.float32(0.09229941),
'ARmax10': np.float32(0.12094776),
'ARmax100': np.float32(0.12880991),
'ARs': np.float32(0.032882884),
'mask_AP': np.float32(0.05688256),
'mask_AP50': np.float32(0.14087473),
'mask_AP75': np.float32(0.03236862),
'mask_APl': np.float32(0.17055447),
'mask_APm': np.float32(0.022371016),
'mask_APs': np.float32(0.0006784558),
'mask_ARl': np.float32(0.2093633),
'mask_ARm': np.float32(0.039542485),
'mask_ARmax1': np.float32(0.06192784),
'mask_ARmax10': np.float32(0.071944),
'mask_ARmax100': np.float32(0.074098006),
'mask_ARs': np.float32(0.0036036037),
'steps_per_second': 4.596630572069489,
'validation_loss': np.float32(0.0)}
Load logs in tensorboard
%load_ext tensorboard
%tensorboard --logdir "./trained_model"
Saving and exporting the trained model
The keras.Model object returned by train_lib.run_experiment expects the data to be normalized by the dataset loader using the same mean and variance statiscics in preprocess_ops.normalize_image(image, offset=MEAN_RGB, scale=STDDEV_RGB). This export function handles those details, so you can pass tf.uint8 images and get the correct results.
export_saved_model_lib.export_inference_graph(
input_type='image_tensor',
batch_size=1,
input_image_size=[HEIGHT, WIDTH],
params=exp_config,
checkpoint_path=tf.train.latest_checkpoint(model_dir),
export_dir=export_dir)
/tmpfs/src/tf_docs_env/lib/python3.11/site-packages/tf_keras/src/engine/functional.py:641: UserWarning: Input dict contained keys ['6'] which did not match any model input. They will be ignored by the model. inputs = self._flatten_to_reference_inputs(inputs) WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.maskrcnn_model.MaskRCNNModel object at 0x7fe8f0349a50>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.maskrcnn_model.MaskRCNNModel object at 0x7fe8f0349a50>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.convolutional.conv2d.Conv2D object at 0x7fe8ac3d16d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.convolutional.conv2d.Conv2D object at 0x7fe8ac3d16d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.normalization.batch_normalization.BatchNormalization object at 0x7fe8605dd150>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.normalization.batch_normalization.BatchNormalization object at 0x7fe8605dd150>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.core.activation.Activation object at 0x7fec7f027a50>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.core.activation.Activation object at 0x7fec7f027a50>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe86057e3d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe86057e3d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fea47e49550>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fea47e49550>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe86057df90>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe86057df90>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe8c856aa10>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe8c856aa10>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe8c85947d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe8c85947d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe8c851e3d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe8c851e3d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe7fc7b42d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <tf_keras.src.layers.merging.add.Add object at 0x7fe7fc7b42d0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.detection_generator.DetectionGenerator object at 0x7fe8f0363590>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.detection_generator.DetectionGenerator object at 0x7fe8f0363590>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.mask_sampler.MaskSampler object at 0x7fe83a112390>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.mask_sampler.MaskSampler object at 0x7fe83a112390>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.roi_sampler.ROISampler object at 0x7fe7ec645010>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.roi_sampler.ROISampler object at 0x7fe7ec645010>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.box_sampler.BoxSampler object at 0x7fe7ec647fd0>, because it is not built. WARNING:tensorflow:Skipping full serialization of TF-Keras layer <official.vision.modeling.layers.box_sampler.BoxSampler object at 0x7fe7ec647fd0>, because it is not built. INFO:tensorflow:Assets written to: ./exported_model/assets INFO:tensorflow:Assets written to: ./exported_model/assets
Inference from Trained Model
def load_image_into_numpy_array(path):
"""Load an image from file into a numpy array.
Puts image into numpy array to feed into tensorflow graph.
Note that by convention we put it into a numpy array with shape
(height, width, channels), where channels=3 for RGB.
Args:
path: the file path to the image
Returns:
uint8 numpy array with shape (img_height, img_width, 3)
"""
image = None
if(path.startswith('http')):
response = urlopen(path)
image_data = response.read()
image_data = BytesIO(image_data)
image = Image.open(image_data)
else:
image_data = tf.io.gfile.GFile(path, 'rb').read()
image = Image.open(BytesIO(image_data))
(im_width, im_height) = image.size
return np.array(image.getdata()).reshape(
(1, im_height, im_width, 3)).astype(np.uint8)
def build_inputs_for_object_detection(image, input_image_size):
"""Builds Object Detection model inputs for serving."""
image, _ = resize_and_crop_image(
image,
input_image_size,
padded_size=input_image_size,
aug_scale_min=1.0,
aug_scale_max=1.0)
return image
Visualize test data
num_of_examples = 3
test_tfrecords = tf.io.gfile.glob('./lvis_tfrecords/val*')
test_ds = tf.data.TFRecordDataset(test_tfrecords).take(num_of_examples)
show_batch(test_ds)
Importing SavedModel
imported = tf.saved_model.load(export_dir)
model_fn = imported.signatures['serving_default']
WARNING:absl:Importing a function (__inference_internal_grad_fn_413497) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418465) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416737) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418285) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415990) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419167) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414316) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414073) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414037) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418942) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418123) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417088) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418924) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415216) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418483) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415675) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416143) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414100) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414955) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416548) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418780) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416530) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413785) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419644) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413596) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419329) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418744) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416134) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413740) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415504) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418573) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413803) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413686) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413956) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416899) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418042) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417196) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414703) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418015) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415945) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417727) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419176) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416710) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413641) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413416) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415423) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416611) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414010) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418159) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414721) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416854) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415135) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418312) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417862) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418213) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417781) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413929) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417007) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414793) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417970) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414523) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414154) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419221) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418996) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415288) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414487) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415810) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418393) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418060) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416989) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418222) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417124) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416242) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414406) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414163) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414064) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415549) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416323) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414757) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418762) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416404) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417619) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419158) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413812) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417628) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414694) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418690) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415234) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413821) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415576) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419212) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416863) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418609) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417826) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417763) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415747) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414325) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413920) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416341) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414478) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414253) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416800) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413731) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416287) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414649) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418915) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417133) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416557) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414397) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417610) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419140) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415441) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414550) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416953) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415180) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413776) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415198) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416926) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417529) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416584) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417421) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416161) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417700) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418789) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415405) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416089) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418447) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418204) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417709) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418168) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414136) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419104) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418618) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416206) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416107) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416350) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416314) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418114) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415243) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415189) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417394) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417898) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418051) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414082) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419680) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415981) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413857) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418330) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416062) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415252) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417052) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416575) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414001) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413632) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416494) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419545) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413407) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418681) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415729) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414334) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416359) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415864) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414964) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415720) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416665) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415693) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416035) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419239) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416278) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416935) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418267) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418231) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417520) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415369) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418420) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415792) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418708) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418555) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415468) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419284) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418906) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419626) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419590) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414595) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418879) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419473) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419455) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414352) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416116) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417502) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418366) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416071) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419599) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414982) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417070) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413533) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415225) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418321) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414505) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416791) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416971) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416782) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413587) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414532) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415009) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416602) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416125) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417952) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414109) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415585) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417313) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416845) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416962) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419446) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415207) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418186) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413560) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415603) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415486) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413452) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416431) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417322) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415882) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418303) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417043) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417565) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419203) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413830) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416629) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416872) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414802) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413758) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414181) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414433) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413380) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416305) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413398) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417664) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415765) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416251) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416503) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418519) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416224) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415963) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414469) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413524) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413389) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419266) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413875) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418987) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419311) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417448) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418672) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416656) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418717) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418456) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416881) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415090) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419572) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414622) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419248) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417142) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413470) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419410) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417475) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418105) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418798) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416980) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418807) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417151) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414739) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417079) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416377) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413794) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418510) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414496) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415999) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414307) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414604) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413479) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418348) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414343) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418402) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413965) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417286) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414388) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418645) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417844) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416053) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418537) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419428) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419356) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414847) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418492) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415594) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415054) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419518) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414919) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415846) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414631) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419671) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417790) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415738) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419068) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419230) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414145) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418897) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417880) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415891) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415639) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416170) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415630) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414190) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415117) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413542) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414019) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413983) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415351) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417943) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413902) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419365) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416467) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415954) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419194) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419383) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414235) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417511) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413668) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418951) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413569) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416368) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418501) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416026) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413434) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414262) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417223) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419149) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414946) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415396) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416008) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415279) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417556) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419347) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414811) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415126) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419419) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419122) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417673) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414838) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413551) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413992) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419464) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419491) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418384) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419086) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417961) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414055) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413695) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418636) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417988) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415936) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419635) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419338) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417997) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415081) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415306) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417646) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417259) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416386) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415855) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414901) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417592) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414127) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414379) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413623) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415450) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417034) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415171) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415099) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419401) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418861) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417349) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415072) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414640) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417160) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417061) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415036) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418474) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418276) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414730) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417331) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415297) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417403) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415684) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418960) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416638) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419689) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418933) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418591) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413659) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414577) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413911) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414775) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414451) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418816) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416764) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416413) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418870) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414874) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415567) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417187) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419581) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418096) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417277) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416476) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417718) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415063) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415774) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415387) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419320) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413488) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418825) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416566) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413650) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414991) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418546) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414370) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416179) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416269) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414973) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416917) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416044) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418654) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419113) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416197) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415153) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418852) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414784) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418411) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418699) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414829) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419059) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419005) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416836) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414568) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414280) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413443) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414613) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416395) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415360) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419653) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413884) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415666) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416998) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416215) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418834) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417745) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418339) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414685) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417484) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418438) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417574) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413713) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418888) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417097) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415324) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417916) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414361) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416332) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418357) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419563) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416512) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418240) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415414) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419662) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414415) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415261) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416080) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417412) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415513) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417241) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417799) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419509) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414289) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416755) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413425) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419131) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419554) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419050) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417907) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416188) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419500) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414514) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416701) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417016) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414244) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416485) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415000) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413767) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416422) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415837) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415927) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419257) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417358) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417439) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419032) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416539) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413974) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417547) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417304) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416449) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418753) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415702) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413677) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417493) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418969) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418132) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416458) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418663) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419392) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415333) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414199) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415909) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417754) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419077) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419293) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418024) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418033) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418150) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418528) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415162) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419095) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414460) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414442) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414910) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419041) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413866) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419374) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416944) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416017) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415783) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413893) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414892) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418564) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418375) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417682) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417808) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417115) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415432) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414586) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415819) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414028) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415045) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413839) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419275) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415918) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419014) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414541) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418006) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417637) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414226) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415477) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415801) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413506) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415342) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415828) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418771) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415459) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418249) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417232) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414424) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416098) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415027) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417772) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417169) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417376) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414658) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418726) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416620) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418069) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416152) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414667) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417601) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416692) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417457) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417025) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416647) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415315) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417889) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413461) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417853) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417691) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418141) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414865) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413722) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416593) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416674) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417268) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415873) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419482) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415711) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418177) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415018) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415657) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417430) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416296) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419536) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418078) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414559) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414937) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418735) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419302) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416719) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415531) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414208) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415270) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418258) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417871) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417340) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416908) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414676) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418429) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415540) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413614) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413704) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415144) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419608) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418600) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416440) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418627) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414091) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414046) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414856) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418582) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413947) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417205) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413605) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417736) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415756) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418294) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414118) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416728) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414271) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419185) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414820) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419527) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418087) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417934) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415612) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419023) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414928) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417925) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415900) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414217) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414766) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413938) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416818) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415378) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414298) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414172) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417979) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415648) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419437) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416890) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416827) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414748) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416773) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415522) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417835) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415621) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417250) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418843) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417214) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416260) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414712) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415972) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417178) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417466) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416746) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418978) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417817) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_414883) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416233) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413578) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417367) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417295) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417385) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417106) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413515) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415108) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416521) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416683) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413848) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_413749) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_419617) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_418195) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417538) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417655) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415495) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_416809) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_417583) with ops with unsaved custom gradients. Will likely fail if a gradient is requested. WARNING:absl:Importing a function (__inference_internal_grad_fn_415558) with ops with unsaved custom gradients. Will likely fail if a gradient is requested.
Visualize predictions
def reframe_image_corners_relative_to_boxes(boxes):
"""Reframe the image corners ([0, 0, 1, 1]) to be relative to boxes.
The local coordinate frame of each box is assumed to be relative to
its own for corners.
Args:
boxes: A float tensor of [num_boxes, 4] of (ymin, xmin, ymax, xmax)
coordinates in relative coordinate space of each bounding box.
Returns:
reframed_boxes: Reframes boxes with same shape as input.
"""
ymin, xmin, ymax, xmax = (boxes[:, 0], boxes[:, 1], boxes[:, 2], boxes[:, 3])
height = tf.maximum(ymax - ymin, 1e-4)
width = tf.maximum(xmax - xmin, 1e-4)
ymin_out = (0 - ymin) / height
xmin_out = (0 - xmin) / width
ymax_out = (1 - ymin) / height
xmax_out = (1 - xmin) / width
return tf.stack([ymin_out, xmin_out, ymax_out, xmax_out], axis=1)
def reframe_box_masks_to_image_masks(box_masks, boxes, image_height,
image_width, resize_method='bilinear'):
"""Transforms the box masks back to full image masks.
Embeds masks in bounding boxes of larger masks whose shapes correspond to
image shape.
Args:
box_masks: A tensor of size [num_masks, mask_height, mask_width].
boxes: A tf.float32 tensor of size [num_masks, 4] containing the box
corners. Row i contains [ymin, xmin, ymax, xmax] of the box
corresponding to mask i. Note that the box corners are in
normalized coordinates.
image_height: Image height. The output mask will have the same height as
the image height.
image_width: Image width. The output mask will have the same width as the
image width.
resize_method: The resize method, either 'bilinear' or 'nearest'. Note that
'bilinear' is only respected if box_masks is a float.
Returns:
A tensor of size [num_masks, image_height, image_width] with the same dtype
as `box_masks`.
"""
resize_method = 'nearest' if box_masks.dtype == tf.uint8 else resize_method
# TODO(rathodv): Make this a public function.
def reframe_box_masks_to_image_masks_default():
"""The default function when there are more than 0 box masks."""
num_boxes = tf.shape(box_masks)[0]
box_masks_expanded = tf.expand_dims(box_masks, axis=3)
resized_crops = tf.image.crop_and_resize(
image=box_masks_expanded,
boxes=reframe_image_corners_relative_to_boxes(boxes),
box_indices=tf.range(num_boxes),
crop_size=[image_height, image_width],
method=resize_method,
extrapolation_value=0)
return tf.cast(resized_crops, box_masks.dtype)
image_masks = tf.cond(
tf.shape(box_masks)[0] > 0,
reframe_box_masks_to_image_masks_default,
lambda: tf.zeros([0, image_height, image_width, 1], box_masks.dtype))
return tf.squeeze(image_masks, axis=3)
input_image_size = (HEIGHT, WIDTH)
plt.figure(figsize=(20, 20))
min_score_thresh = 0.40 # Change minimum score for threshold to see all bounding boxes confidences
for i, serialized_example in enumerate(test_ds):
plt.subplot(1, 3, i+1)
decoded_tensors = tf_ex_decoder.decode(serialized_example)
image = build_inputs_for_object_detection(decoded_tensors['image'], input_image_size)
image = tf.expand_dims(image, axis=0)
image = tf.cast(image, dtype = tf.uint8)
image_np = image[0].numpy()
result = model_fn(image)
# Visualize detection and masks
if 'detection_masks' in result:
# we need to convert np.arrays to tensors
detection_masks = tf.convert_to_tensor(result['detection_masks'][0])
detection_boxes = tf.convert_to_tensor(result['detection_boxes'][0])
detection_masks_reframed = reframe_box_masks_to_image_masks(
detection_masks, detection_boxes/256.0,
image_np.shape[0], image_np.shape[1])
detection_masks_reframed = tf.cast(
detection_masks_reframed > min_score_thresh,
np.uint8)
result['detection_masks_reframed'] = detection_masks_reframed.numpy()
visualization_utils.visualize_boxes_and_labels_on_image_array(
image_np,
result['detection_boxes'][0].numpy(),
(result['detection_classes'][0] + 0).numpy().astype(int),
result['detection_scores'][0].numpy(),
category_index=category_index,
use_normalized_coordinates=False,
max_boxes_to_draw=200,
min_score_thresh=min_score_thresh,
instance_masks=result.get('detection_masks_reframed', None),
line_thickness=4)
plt.imshow(image_np)
plt.axis("off")
plt.show()
