README

The pytorch code of some common neural network models, running on MNIST dataet.

Requirements

Python >=3.8

pip install -r requirements.txt

Dataset

MNIST

• num of train data : num of valid data = 9 : 1

Models

• LeNet: 《Gradient-Based Learning Applied to Document Recognition》
• AlexNet: 《ImageNet classification with deep convolutional neural networks 》
• ResNet: 《Deep Residual Learning for Image Recognition》
• DenseNet: 《Densely Connected Convolutional Networks》
• MobileNet：《MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications》

Instructions

Run inference：

python inference.py \
--model AlexNet \
--lr 0.005 \
--dropout 0.5 \
--batchsize 64 \
--num_epochs 10

The output will be written to log\$model_name.txt

Add new model or dataset

• Add new model : Define new model class in model.py
• Add new dataset : Define new Dataset and Dataloader class in utils.py