Skip to content

Latest commit

 

History

History
39 lines (39 loc) · 616 Bytes

File metadata and controls

39 lines (39 loc) · 616 Bytes

model

  • autoencoder
    • Stacked AutoEncoder
    • Vanilla autoencoder
    • Multilayer autoencoder
    • Convolutional autoencoder
    • Regularized autoencoder
      • denoising autoencoder
      • sparse autoencoder
        训练不易收敛,逐层训练可以缓解。
        
  • sgnet
  • xlnet
  • GPT
  • ELMO
  • crcnn
  • memory network
    • e2emn
    • dmn
  • attention
    • structure att
    • global/local att
    • soft/hard att
  • gan
    • seqgan
    • wgan
  • seq2seq
    • transformer
  • tree lstm
    • todo
  • nlg
    • MaliGAN,RankGAN,DP-GAN和LeakGAN
  • gnn
    • diffpool
  • text cluster
    • atomic clusters
  • capsnet
    • a