• AlexNet Description
• Dataset
• Environment Requirements
• Model Source
• Quick Start

AlexNet Description

AlexNet was proposed in 2012, one of the most influential neural networks. It got big success in ImageNet Dataset recognition than other models.

Paper: Krizhevsky A, Sutskever I, Hinton G E. ImageNet Classification with Deep ConvolutionalNeural Networks. Advances In Neural Information Processing Systems. 2012.

Dataset

Dataset used: CIFAR-10

• Dataset size：175M，60,000 32*32 colorful images in 10 classes
  • Train：146M，50,000 images
  • Test：29.3M，10,000 images

Environment Requirements

• Hardware（Ascend/GPU/CPU）

Model Source

In this case, the source of the model definition or training script corresponding to the implementation of PyTorch is as follows:

https://github.com/pytorch/vision/blob/main/torchvision/models/alexnet.py

Quick Start

This case shows the AlexNet model implemented by pytorch and the version converted based on MSAadpter, you can start training and evaluation as follows:

# PyTorch
# enter script dir, train AlexNet by PyTorch
python alexnet_torch.py [--mode 'train'] [--device 'cpu'|'cuda'] [--epoch] [--save_path]
# example: python alexnet_torch.py --device cpu --save_path ./alexnet.pth

# enter script dir, evaluate AlexNet by PyTorch
python alexnet_torch.py [--mode 'test'] [--device 'cpu'|'cuda'] [--load_path]
# example: python alexnet_torch.py --mode test --load_path ./alexnet.pth

# MSAdapter.pytorch
# enter script dir, train AlexNet by MSAdapter.pytorch
python alexnet_adapter.py [--mode 'train'] [--device 'CPU'|'GPU'|'ASCEND'] [--epoch] [--save_path]
# example: python alexnet_adapter.py --device GPU --save_path ./alexnet.ckpt

# enter script dir, evaluate AlexNet by MSAdapter.pytorch
python alexnet_adapter.py [--mode 'test'] [--device 'CPU'|'GPU'|'ASCEND'] [--load_path]
# example: python alexnet_adapter.py --mode test --load_path ./alexnet.ckpt