warmspringwinds / pytorch-image-classification

Image classification in pytorch on Cifar and Imagenet

pytorch-image-classification

Image classification in pytorch on Cifar and Imagenet

Cifar-10

Results

Model	Test data	Our Accuracy	Original Accuracy	Related paper
Resnet-20	Cifar-10	0.9213	0.9125	Deep Residual Learning

Imagenet

Data preparation

1. sudo apt-get install aria2

2. Navigate to http://academictorrents.com/collection/imagenet-2012 and download torrent files for train and validation datasets.

3. Follow this page https://aria2.github.io/ and do aria2c path_to_you_torrent_file.torrent for both validation and training files.

4. By now, you should have two files in your folder ILSVRC2012_img_train.tar and ILSVRC2012_img_val.tar.

5. Run this command:

mkdir train && mv ILSVRC2012_img_train.tar train/ && cd train
tar -xvf ILSVRC2012_img_train.tar && rm -f ILSVRC2012_img_train.tar
find . -name "*.tar" | while read NAME ; do mkdir -p "${NAME%.tar}"; tar -xvf "${NAME}" -C "${NAME%.tar}"; rm -f "${NAME}"; done
cd ..

6. And this:

mkdir val && mv ILSVRC2012_img_val.tar val/ && cd val && tar -xvf ILSVRC2012_img_val.tar
wget -qO- https://raw.githubusercontent.com/soumith/imagenetloader.torch/master/valprep.sh | bash

Credits:

• https://github.com/facebook/fb.resnet.torch/blob/master/INSTALL.md#download-the-imagenet-dataset
• https://github.com/pytorch/examples/tree/e0d33a69bec3eb4096c265451dbb85975eb961ea/imagenet
• https://aria2.github.io/
• https://martin-thoma.com/download-data/