InceptionV4-PyTorch

Overview

This repository contains an op-for-op PyTorch reimplementation of Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning.

Table of contents

• InceptionV4-PyTorch
  • Overview
  • Table of contents
  • Download weights
  • Download datasets
  • How Test and Train
    • Test
    • Train model
    • Resume train model
  • Result
  • Contributing
  • Credit
    • Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

Download weights

• Google Driver
• Baidu Driver

Download datasets

Contains MNIST, CIFAR10&CIFAR100, TinyImageNet_200, MiniImageNet_1K, ImageNet_1K, Caltech101&Caltech256 and more etc.

• Google Driver
• Baidu Driver

Please refer to README.md in the data directory for the method of making a dataset.

How Test and Train

Both training and testing only need to modify the config.py file.

Test

• line 29: model_arch_name change to inception_v4.
• line 31: model_mean_parameters change to [0.485, 0.456, 0.406].
• line 32: model_std_parameters change to [0.229, 0.224, 0.225].
• line 34: model_num_classes change to 1000.
• line 36: mode change to test.
• line 89: model_weights_path change to ./results/pretrained_models/InceptionV4-ImageNet_1K-2069673f.pth.tar.

python3 test.py

Train model

• line 29: model_arch_name change to inception_v4.
• line 31: model_mean_parameters change to [0.485, 0.456, 0.406].
• line 32: model_std_parameters change to [0.229, 0.224, 0.225].
• line 34: model_num_classes change to 1000.
• line 36: mode change to train.
• line 50: pretrained_model_weights_path change to ./results/pretrained_models/InceptionV4-ImageNet_1K-2069673f.pth.tar.

python3 train.py

Resume train model

• line 29: model_arch_name change to inception_v4.
• line 31: model_mean_parameters change to [0.485, 0.456, 0.406].
• line 32: model_std_parameters change to [0.229, 0.224, 0.225].
• line 34: model_num_classes change to 1000.
• line 36: mode change to train.
• line 53: resume change to ./samples/inception_v4-ImageNet_1K/epoch_xxx.pth.tar.

python3 train.py

Result

Source of original paper results: https://arxiv.org/pdf/1602.07261v2.pdf)

In the following table, the top-x error value in () indicates the result of the project, and - indicates no test.

Model	Dataset	Top-1 error (val)	Top-5 error (val)
inception_v4	ImageNet_1K	20.0%(22.2%)	5.0%(6.1%)
inception_v4-resnet_v2	ImageNet_1K	19.9%(23.5%)	4.9%(6.6%)

# Download `InceptionV4-ImageNet_1K-2069673f.pth.tar` weights to `./results/pretrained_models`
# More detail see `README.md<Download weights>`
python3 ./inference.py 

Input:

Output:

Build `inception_v4` model successfully.
Load `inception_v4` model weights `/InceptionV4-PyTorch/results/pretrained_models/InceptionV4-ImageNet_1K-2069673f.pth.tar` successfully.
tench, Tinca tinca                                                          (85.94%)
barracouta, snoek                                                           (1.43%)
reel                                                                        (0.18%)
gar, garfish, garpike, billfish, Lepisosteus osseus                         (0.16%)
goldfish, Carassius auratus                                                 (0.09%)

Contributing

If you find a bug, create a GitHub issue, or even better, submit a pull request. Similarly, if you have questions, simply post them as GitHub issues.

I look forward to seeing what the community does with these models!

Credit

Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke, Alex Alemi

Abstract

Convolutional networks are at the core of most state-of-the-art computer vision solutions for a wide variety of tasks. Since 2014 very deep convolutional networks started to become mainstream, yielding substantial gains in various benchmarks. Although increased model size and computational cost tend to translate to immediate quality gains for most tasks (as long as enough labeled data is provided for training), computational efficiency and low parameter count are still enabling factors for various use cases such as mobile vision and big-data scenarios. Here we are exploring ways to scale up networks in ways that aim at utilizing the added computation as efficiently as possible. We benchmark our methods on the ILSVRC 2012 classification challenge validation set and demonstrate substantial gains over the state of the art via to carefully factorized convolutions and aggressive regularization: 21.2% top-1 and 5.6% top-5 error for single frame evaluation using a network with a computational cost of 5 billion multiply-adds per inference and with using less than 25 million parameters.

[Paper]

@inproceedings{szegedy2016rethinking,
  title={Rethinking the inception architecture for computer vision},
  author={Szegedy, Christian and Vanhoucke, Vincent and Ioffe, Sergey and Shlens, Jon and Wojna, Zbigniew},
  booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
  pages={2818--2826},
  year={2016}
}