tatp22/linformer-pytorch

Issues

• Question about how to modify to predict on a series of sparse number

  #27 opened 8 months ago by JonasLi-19
  0

• Question: Is Linformer permutation equivariant (set-operation)?

  #26 opened 2 years ago by nmakes
  5

• Loss goes to 0 when using LinformerLM

  #25 opened 3 years ago by terencenwz
  2

• Different number of tokens and Character Level Modeling

  #24 opened 4 years ago by wajihullahbaig
  2

• causal_mask of the decoder

  #16 opened 4 years ago by burcehan
  4

• Error when using method="no_params" and GPU, because E and F incorrectly remain on CPU

  #20 opened 4 years ago by RaivoKoot
  4

• Error with DistributedDataParallel and parameter_sharing="layerwise"

  #23 opened 4 years ago by blizda
  2

• Error with DistributedDataParallel

  #22 opened 4 years ago by blizda
  2

• Would you like to release the pretrain tutorial?

  #4 opened 4 years ago by RyanHuangNLP
  11

• embeddings_mask datatype

  #14 opened 4 years ago by tongcu
  1

• How to interpret the visualization results?

  #15 opened 4 years ago by mertyyanik
  2

• Possible bug

  #8 opened 4 years ago by tatp22
  2

• Any result on any benchmark?

  #13 opened 4 years ago by twangnh
  5

• padding mask and attention mask

  #11 opened 4 years ago by zackchen-lb
  10

• Enquiry about your implementation

  #10 opened 4 years ago by riven314
  2

• Huggingface

  #3 opened 4 years ago by flozi00
  3

• Composed linear layers?

  #6 opened 4 years ago by apeguero1
  5

• Any performance test on different checkpoint level ?

  #5 opened 4 years ago by phongnhhn92
  2

• input seg length

  #2 opened 4 years ago by xinqipony
  1

• Will any pretrained linformer models be open sourced?

  #1 opened 4 years ago by pchankh
  1