Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/facebookarchive/fb.resnet.torch
Torch implementation of ResNet from http://arxiv.org/abs/1512.03385 and training scripts
https://github.com/facebookarchive/fb.resnet.torch
Last synced: about 2 months ago
JSON representation
Torch implementation of ResNet from http://arxiv.org/abs/1512.03385 and training scripts
- Host: GitHub
- URL: https://github.com/facebookarchive/fb.resnet.torch
- Owner: facebookarchive
- License: other
- Archived: true
- Created: 2016-01-12T20:29:16.000Z (almost 9 years ago)
- Default Branch: master
- Last Pushed: 2022-08-24T15:23:59.000Z (about 2 years ago)
- Last Synced: 2024-05-23T07:43:47.219Z (6 months ago)
- Language: Lua
- Size: 97.7 KB
- Stars: 2,277
- Watchers: 120
- Forks: 664
- Open Issues: 60
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
Archived
========
This project is no longer maintained. Consider using [ImageNet training in PyTorch](https://github.com/pytorch/examples/tree/master/imagenet) instead.
ResNet training in Torch
============================
This implements training of residual networks from [Deep Residual Learning for Image Recognition](http://arxiv.org/abs/1512.03385) by Kaiming He, et. al.
[We wrote a more verbose blog post discussing this code, and ResNets in general here.](http://torch.ch/blog/2016/02/04/resnets.html)
## Requirements
See the [installation instructions](INSTALL.md) for a step-by-step guide.
- Install [Torch](http://torch.ch/docs/getting-started.html) on a machine with CUDA GPU
- Install [cuDNN v4 or v5](https://developer.nvidia.com/cudnn) and the Torch [cuDNN bindings](https://github.com/soumith/cudnn.torch/tree/R4)
- Download the [ImageNet](http://image-net.org/download-images) dataset and [move validation images](https://github.com/facebook/fb.resnet.torch/blob/master/INSTALL.md#download-the-imagenet-dataset) to labeled subfolders
If you already have Torch installed, update `nn`, `cunn`, and `cudnn`.
## Training
See the [training recipes](TRAINING.md) for addition examples.
The training scripts come with several options, which can be listed with the `--help` flag.
```bash
th main.lua --help
```
To run the training, simply run main.lua. By default, the script runs ResNet-34 on ImageNet with 1 GPU and 2 data-loader threads.
```bash
th main.lua -data [imagenet-folder with train and val folders]
```
To train ResNet-50 on 4 GPUs:
```bash
th main.lua -depth 50 -batchSize 256 -nGPU 4 -nThreads 8 -shareGradInput true -data [imagenet-folder]
```
## Trained models
Trained ResNet 18, 34, 50, 101, 152, and 200 models are [available for download](pretrained). We include instructions for [using a custom dataset](pretrained/README.md#fine-tuning-on-a-custom-dataset), [classifying an image and getting the model's top5 predictions](pretrained/README.md#classification), and for [extracting image features](pretrained/README.md#extracting-image-features) using a pre-trained model.
The trained models achieve better error rates than the [original ResNet models](https://github.com/KaimingHe/deep-residual-networks).
#### Single-crop (224x224) validation error rate
| Network | Top-1 error | Top-5 error |
| ------------- | ----------- | ----------- |
| ResNet-18 | 30.43 | 10.76 |
| ResNet-34 | 26.73 | 8.74 |
| ResNet-50 | 24.01 | 7.02 |
| ResNet-101 | 22.44 | 6.21 |
| ResNet-152 | 22.16 | 6.16 |
| ResNet-200 | 21.66 | 5.79 |
## Notes
This implementation differs from the ResNet paper in a few ways:
**Scale augmentation**: We use the [scale and aspect ratio augmentation](datasets/transforms.lua#L130) from [Going Deeper with Convolutions](http://arxiv.org/abs/1409.4842), instead of [scale augmentation](datasets/transforms.lua#L113) used in the ResNet paper. We find this gives a better validation error.
**Color augmentation**: We use the photometric distortions from [Andrew Howard](http://arxiv.org/abs/1312.5402) in addition to the [AlexNet](http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf)-style color augmentation used in the ResNet paper.
**Weight decay**: We apply weight decay to all weights and biases instead of just the weights of the convolution layers.
**Strided convolution**: When using the bottleneck architecture, we use stride 2 in the 3x3 convolution, instead of the first 1x1 convolution.