pytorch-classification

This repository is a fork of pytorch-classification, extended with new features.

Features

• Unified interface for different network architectures
• Multi-GPU support
• Training progress bar with rich info
• Training log and training curve visualization code (see ./utils/logger.py)
• New: Grad-CAM visualization script (gradcam.py) compatible with all supported models.
• New: Expanded model support for EfficientNet, ViT, ConvNeXt, and more.

Install

• Install PyTorch
• Clone recursively

git clone --recursive https://github.com/adarsh-crafts/pytorch-classification-extended.git

Data

Organize your dataset in an ImageNet-style directory structure as follows:

data/
│
├── train/
│   ├── class 1/
│   ├── class 2/
│   ├── class 3/
│
├── val/
│   ├── class 1/
│   ├── class 2/
│   ├── class 3/

Training

Please see the Training recipes for how to train the models.

Visualization (Grad-CAM)

A universal Grad-CAM script (gradcam.py) is included for model visualization. It is compatible with all supported architectures (including ViT) and loads checkpoints saved by the training script (train.py).

Example Usage

python gradcam.py \
    --checkpoint /path/to/checkpoint/model_best.pth.tar \
    --image /path/to/your/image.jpg \
    --arch resnet50 \
    --num-classes 1000 \
    --output gradcam_output.png

Training Results

CIFAR

Top1 error rate on the CIFAR-10/100 benchmarks are reported. You may get different results when training your models with different random seed. Note that the number of parameters are computed on the CIFAR-10 dataset.

Model	Params (M)	CIFAR-10 (%)	CIFAR-100 (%)
alexnet	2.47	22.78	56.13
vgg19_bn	20.04	6.66	28.05
ResNet-110	1.70	6.11	28.86
PreResNet-110	1.70	4.94	23.65
WRN-28-10 (drop 0.3)	36.48	3.79	18.14
ResNeXt-29, 8x64	34.43	3.69	17.38
ResNeXt-29, 16x64	68.16	3.53	17.30
DenseNet-BC (L=100, k=12)	0.77	4.54	22.88
DenseNet-BC (L=190, k=40)	25.62	3.32	17.17

ImageNet

Single-crop (224x224) validation error rate is reported.

Model	Params (M)	Top-1 Error (%)	Top-5 Error (%)
ResNet-18	11.69	30.09	10.78
ResNeXt-50 (32x4d)	25.03	22.6	6.29

Pretrained models

Our trained models and training logs are downloadable at OneDrive.

Supported Architectures

CIFAR-10 / CIFAR-100

Since the size of images in CIFAR dataset is 32x32, popular network structures for ImageNet need some modifications to adapt this input size. The modified models is in the package models.cifar:

• AlexNet
• VGG (Imported from pytorch-cifar)
• ResNet
• Pre-act-ResNet
• ResNeXt (Imported from ResNeXt.pytorch)
• Wide Residual Networks (Imported from WideResNet-pytorch)
• DenseNet

ImageNet

• All models in torchvision.models (e.g., alexnet, vgg, resnet, densenet, inception_v3, squeezenet, mobilenet, efficientnet, vit, convnext, etc.)
• ResNeXt (Custom implementation via models.imagenet)
• Wide Residual Networks

Contribute

Feel free to create a pull request if you find any bugs or you want to contribute (e.g., more datasets and more network structures).