mit-han-lab/lite-transformer

Issues

• could you share the tensorboard log file? thank you so much!

  #5 opened 5 months ago by luogen1996
  1

• Export model to ONNX

  #24 opened 5 months ago by suyuzhang
  1

• Please share your quantization, quantization+pruning checkpoints

  #26 opened 5 months ago by kishorepv
  2

• Can not get the result as the paper if train the transformer from scratch.

  #30 opened 5 months ago by tomshalini
  3

• about the global and local features in fig 3

  #35 opened 5 months ago by sanwei111
  3

• TransformerEncoderLayer

  #36 opened 5 months ago by sanwei111
  5

• wmt16_en_de dataset link

  #41 opened 5 months ago by topbookcc
  2

• model pruning

  #42 opened 5 months ago by AIikai
  1

• Can‘t find the cnn branch,

  #43 opened 5 months ago by gwyanCN
  1

• about kernel size

  #37 opened 3 years ago by sanwei111
  1

• About data !

  #40 opened 3 years ago by veryhigh
  1

• about dynamicconv_cuda

  #38 opened 3 years ago by sanwei111
  1

• about padding！！！

  #39 opened 3 years ago by sanwei111
  2

• Model size confuse

  #12 opened 4 years ago by zml24
  1

• in paragra 4 of

  #33 opened 4 years ago by sanwei111
  1

• in the paragra 4 of paper

  #34 opened 4 years ago by sanwei111
  1

• How to measure the FLOPs/MACs?

  #31 opened 4 years ago by ranery
  2

• Error while testing the model

  #29 opened 4 years ago by tomshalini
  8

• Error while evaluating model

  #25 opened 4 years ago by kishorepv
  9

• Missing Data Preparation section for the CNN / DailyMail dataset

  #28 opened 4 years ago by cronopioelectronico
  1

• Quantization

  #22 opened 4 years ago by zilunpeng
  1

• transfomer model with different paramters

  #23 opened 4 years ago by ChuanyangZheng
  3

• What functions are achieved in the cu code? The cu code is too hard for me to understand. Thank you.

  #21 opened 4 years ago by guotong1988
  1

• Why do you recode the cpp code and cu code? What function is necessary?

  #20 opened 4 years ago by guotong1988
  1

• Could you please point out the core code, as there are too many fairseq code. Thank you!

  #19 opened 4 years ago by guotong1988
  2

• Will you release the TensorFlow code in the future?

  #18 opened 4 years ago by guotong1988
  1

• CNN\DM dateset preprocess (bpe 30K)

  #17 opened 4 years ago by Wangt-CN
  1

• wmt14 en-fr data processing problem

  #16 opened 4 years ago by macn3388
  3

• Is there any link for downloading iwslt14.de-en pretrained model?

  #15 opened 4 years ago by macn3388
  1

• Is there any guidance on preparing the cnndm dataset?

  #14 opened 4 years ago by macn3388
  1

• Data preprocessing

  #13 opened 4 years ago by swgu98
  1

• Model Compression

  #4 opened 4 years ago by kalyangvs
  5

• Wrong key_padding_mask position?

  #6 opened 4 years ago by godweiyang
  3

• Could you share Quantify and Pruning script? Thank you very much!

  #11 opened 4 years ago by fansiawang
  2

• DeprecationWarning

  #7 opened 4 years ago
  1

• register_task of abstractive summarization

  #8 opened 4 years ago by zhangyongwei756
  1

• training config for wikitext103

  #2 opened 5 years ago by pichuang1984
  4

• Applying factorized embedding

  #3 opened 5 years ago by asharma20
  4

• Summarization checkpoint release ?

  #1 opened 5 years ago by astariul
  1