Record my favorite and useful deep learning models encountered in the learning process, including but not limited to classic convolutional neural networks (e.g. AlexNet,ResNet), attention-mechanism related modules, transformer related models (e.g. ViT,Efficient ViT), etc.

I will continue to update more models as I learn. All models are based on the pytorch implementation and use cuda version 11.8. Another important point is that a large number of implemented models are available in the pytorch framework and transformers library. Reading the source code will improve your coding skills.

The model code has been updated:

• image_segmentation：
  • about_unet：
    • UNet
    • UNet_pp
• about_transformer:
  • Transformer
• ViT_model
  • mobile_vit
• about_attention：
  • MSCAAttention
  • CBAM
• classic_conv：
  • AlexNet
  • SENet
• about_interview
  • Attention
  • Embedding
  • FFN
  • LayerNorm