- বর্ণনা :
ImageNet-PI হল স্ট্যান্ডার্ড ILSVRC2012 ImageNet ডেটাসেটের একটি পুনঃলেবেল করা সংস্করণ যেখানে লেবেলগুলি 16টি গভীর নিউরাল নেটওয়ার্কের সংগ্রহ দ্বারা সরবরাহ করা হয় যার বিভিন্ন আর্কিটেকচার স্ট্যান্ডার্ড ILSVRC2012-তে প্রাক-প্রশিক্ষিত। বিশেষ করে, প্রাক-প্রশিক্ষিত মডেলগুলি tf.keras.applications থেকে ডাউনলোড করা হয়।
নতুন লেবেলগুলি ছাড়াও, ImageNet-PI মডেলগুলির লেবেলে থাকা আত্মবিশ্বাসের আকারে টীকা প্রক্রিয়া সম্পর্কে মেটা-ডেটা এবং প্রতিটি মডেল সম্পর্কে অতিরিক্ত তথ্য সরবরাহ করে।
আরও তথ্যের জন্য দেখুন: ImageNet-PI
হোমপেজ : https://github.com/google-research-datasets/imagenet_pi/
সোর্স কোড :
tfds.datasets.imagenet_pi.Builderসংস্করণ :
-
1.0.0(ডিফল্ট): প্রাথমিক প্রকাশ।
-
ডাউনলোডের আকার :
Unknown sizeডেটাসেটের আকার :
Unknown sizeম্যানুয়াল ডাউনলোডের নির্দেশাবলী : এই ডেটাসেটের জন্য আপনাকে
download_config.manual_dir(ডিফল্ট~/tensorflow_datasets/downloads/manual/) তে সোর্স ডেটা ম্যানুয়ালি ডাউনলোড করতে হবে:
manual_dir-এ দুটি ফাইল থাকা উচিত: ILSVRC2012_img_train.tar এবং ILSVRC2012_img_val.tar। ডেটাসেট ডাউনলোড করার লিঙ্ক পেতে আপনাকে http://www.image-net.org/download-images- এ নিবন্ধন করতে হবে।অটো-ক্যাশেড ( ডকুমেন্টেশন ): অজানা
বিভক্ত :
| বিভক্ত | উদাহরণ |
|---|
- বৈশিষ্ট্য কাঠামো :
FeaturesDict({
'annotator_confidences': Tensor(shape=(16,), dtype=float32),
'annotator_labels': Tensor(shape=(16,), dtype=int64),
'clean_label': ClassLabel(shape=(), dtype=int64, num_classes=1000),
'file_name': Text(shape=(), dtype=string),
'image': Image(shape=(None, None, 3), dtype=uint8),
})
- বৈশিষ্ট্য ডকুমেন্টেশন :
| বৈশিষ্ট্য | শ্রেণী | আকৃতি | টাইপ | বিবরণ |
|---|---|---|---|---|
| বৈশিষ্ট্যঅভিধান | ||||
| অ্যানোটেটর_কনফিডেন্স | টেনসর | (১৬,) | float32 সম্পর্কে | |
| অ্যানোটেটর_লেবেল | টেনসর | (১৬,) | int64 সম্পর্কে | |
| ক্লিন_লেবেল | ক্লাসলেবেল | int64 সম্পর্কে | ||
| ফাইলের নাম | টেক্সট | স্ট্রিং | ||
| চিত্র | ভাবমূর্তি | (কোনটিই নয়, কেউই নয়, ৩) | uint8 |
তত্ত্বাবধানে থাকা কী (
as_supervisedডক দেখুন):('image', 'annotator_labels')চিত্র ( tfds.show_examples ): সমর্থিত নয়।
উদাহরণ ( tfds.as_dataframe ): অনুপস্থিত।
উদ্ধৃতি :
@inproceedings{tram,
author = {Mark Collier and
Rodolphe Jenatton and
Effrosyni Kokiopoulou and
Jesse Berent},
editor = {Kamalika Chaudhuri and
Stefanie Jegelka and
Le Song and
Csaba Szepesv{\'{a} }ri and
Gang Niu and
Sivan Sabato},
title = {Transfer and Marginalize: Explaining Away Label Noise with Privileged
Information},
booktitle = {International Conference on Machine Learning, {ICML} 2022, 17-23 July
2022, Baltimore, Maryland, {USA} },
series = {Proceedings of Machine Learning Research},
volume = {162},
pages = {4219--4237},
publisher = { {PMLR} },
year = {2022},
url = {https://proceedings.mlr.press/v162/collier22a.html},
}
@article{ILSVRC15,
Author = {Olga Russakovsky and Jia Deng and Hao Su and Jonathan Krause and Sanjeev Satheesh and Sean Ma and Zhiheng Huang and Andrej Karpathy and Aditya Khosla and Michael Bernstein and Alexander C. Berg and Li Fei-Fei},
Title = { {ImageNet Large Scale Visual Recognition Challenge} },
Year = {2015},
journal = {International Journal of Computer Vision (IJCV)},
doi = {10.1007/s11263-015-0816-y},
volume={115},
number={3},
pages={211-252}
}