cifar10_n

  • Deskripsi :

Versi CIFAR-10 yang diberi label ulang dengan kesalahan anotasi manusia nyata. Untuk setiap pasangan (gambar, label) dalam set kereta CIFAR-10 asli, ini menyediakan beberapa label tambahan yang diberikan oleh anotator manusia asli.

  • Beranda : https://ucsc-real.soe.ucsc.edu:1995/Home.html/

  • Kode sumber : tfds.image_classification.cifar10_n.Cifar10N

  • Versi :

    • 1.0.0 : Rilis awal.
    • 1.0.1 : Memperbaiki salah ketik di kunci worse_label .
    • 1.0.2 : Memperbaiki korespondensi antara anotasi dan gambar.
    • 1.0.3 : Memperbaiki file di MANUAL_DIR .
    • 1.0.4 (default): Memperbaiki pemuatan informasi samping.
  • Ukuran unduhan : 162.17 MiB

  • Ukuran dataset : 147.91 MiB

  • Instruksi pengunduhan manual : Kumpulan data ini mengharuskan Anda mengunduh data sumber secara manual ke download_config.manual_dir (default ke ~/tensorflow_datasets/downloads/manual/ ):
    Unduh 'side_info_cifar10N.csv', 'CIFAR-10_human_ordered.npy' dan 'image_order_c10.npy' dari https://github.com/UCSC-REAL/cifar-10-100n

Kemudian ubah 'CIFAR-10_human_ordered.npy' menjadi file CSV 'CIFAR-10_human_annotations.csv'. Ini dapat dilakukan dengan kode berikut:

import numpy as np
from tensorflow_datasets.core.utils.lazy_imports_utils import pandas as pd
from tensorflow_datasets.core.utils.lazy_imports_utils import tensorflow as tf

human_labels_np_path = '<local_path>/CIFAR-10_human_ordered.npy'
human_labels_csv_path = '<local_path>/CIFAR-10_human_annotations.csv'

with tf.io.gfile.GFile(human_labels_np_path, "rb") as f:
  human_annotations = np.load(f, allow_pickle=True)

df = pd.DataFrame(human_annotations[()])

with tf.io.gfile.GFile(human_labels_csv_path, "w") as f:
  df.to_csv(f, index=False)
Membelah Contoh
'test' 10.000
'train' 50.000
  • Struktur fitur :
FeaturesDict({
    'aggre_label': ClassLabel(shape=(), dtype=int64, num_classes=10),
    'id': Text(shape=(), dtype=string),
    'image': Image(shape=(32, 32, 3), dtype=uint8),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=10),
    'random_label1': ClassLabel(shape=(), dtype=int64, num_classes=10),
    'random_label2': ClassLabel(shape=(), dtype=int64, num_classes=10),
    'random_label3': ClassLabel(shape=(), dtype=int64, num_classes=10),
    'worker1_id': int64,
    'worker1_time': float32,
    'worker2_id': int64,
    'worker2_time': float32,
    'worker3_id': int64,
    'worker3_time': float32,
    'worse_label': ClassLabel(shape=(), dtype=int64, num_classes=10),
})
  • Dokumentasi fitur :
Fitur Kelas Membentuk Dtype Keterangan
fiturDict
aggre_label LabelKelas int64
Indo Teks rangkaian
gambar Gambar (32, 32, 3) uint8
label LabelKelas int64
label_acak1 LabelKelas int64
label_acak2 LabelKelas int64
label_acak3 LabelKelas int64
pekerja1_id Tensor int64
pekerja1_waktu Tensor float32
pekerja2_id Tensor int64
pekerja2_waktu Tensor float32
pekerja3_id Tensor int64
pekerja3_waktu Tensor float32
label_buruk LabelKelas int64

Visualisasi

  • Kutipan :
@inproceedings{wei2022learning,
  title={Learning with Noisy Labels Revisited: A Study Using Real-World Human
  Annotations},
  author={Jiaheng Wei and Zhaowei Zhu and Hao Cheng and Tongliang Liu and Gang
  Niu and Yang Liu},
  booktitle={International Conference on Learning Representations},
  year={2022},
  url={https://openreview.net/forum?id=TBWA6PLJZQm}
}