facebookresearch/FiD

Issues

• Distributed computing (eg, multi-GPU) support

  #13 opened 3 years ago by erensezener
  7

• I've solved it

  #40 opened 8 months ago by E2320107
  0

• Applying beam search to Fusion-in-Decoder

  #39 opened 10 months ago by AyeshGK
  0

• Question about optimizer.pth.tar

  #38 opened a year ago by wzy999hello
  0

• Question about the retriever training

  #30 opened a year ago by szhang42
  1

• question about reported result on NQ dataset

  #3 opened 4 years ago by Rosarubu
  2

• Dataset about squad

  #37 opened a year ago by Xnhyacinth
  0

• Hyper-params for reprodcuing results of base reader in FiD

  #16 opened 3 years ago by akhilkedia
  3

• Why using 100 passages?

  #35 opened a year ago by ShiyuNee
  0

• Data pre-processing technique

  #34 opened a year ago by Cezhann-ops
  0

• @fabrahman Could you provide an update on this issue? I have exactly the same issue, and I found that the code freezes without any error message after executing [line 194 of train_reader.py](https://github.com/facebookresearch/FiD/blob/main/train_reader.py#L194)

  #33 opened a year ago by szh-max
  0

• ModuleNotFoundError

  #32 opened a year ago by Xnhyacinth
  4

• test.py for cross-attention scores is missing

  #24 opened 2 years ago by lixsh6
  2

• Output_scores in generate() method

  #21 opened 2 years ago by nrjvarshney
  1

• Questions for reproducing results

  #8 opened 3 years ago by huvunvidia
  1

• How exactly do you pool the output of the encoder?

  #28 opened 2 years ago by zhuobinggang
  0

• If there are questions with no answer in retrieved data for reader training, will these questions be discarded?

  #27 opened 2 years ago by guzy0324
  0

• CUDA memory suddenly run out of space when only used a quarter of memory

  #26 opened 2 years ago by khang-nguyen2907
  0

• Test

  #25 opened 2 years ago by danielfleischer
  0

• Question about changing LM model to GPT

  #23 opened 2 years ago by daje0601
  0

• Memory leak for dataloader

  #20 opened 3 years ago by jumxglhf
  1

• Tutorial about using pretrained models

  #7 opened 3 years ago by infopg
  2

• Fine-tuning using pretrained model

  #6 opened 3 years ago by SebastianLarssonDTU
  1

• How can I get a checkpoint file of the FiD reader?

  #15 opened 3 years ago by LzyFischer
  1

• Remove this line

  #18 opened 3 years ago
  0

• Question about data loading

  #12 opened 3 years ago by PlusRoss
  1

• Making the code works with BART generator

  #14 opened 3 years ago by fabrahman
  0

• A question about ``passages_index''

  #10 opened 3 years ago by chuzhumin98
  1

• About the hyperparameters of finetuning t5-base

  #11 opened 3 years ago by shunyuzh
  1

• Running FiD with fp16 precision

  #9 opened 3 years ago by huvunvidia
  1

• Support transformers > 4.1.x and tutorials about using the pretrained model?

  #5 opened 4 years ago by velocityCavalry
  1

• Paths to checkpoints, data?

  #1 opened 4 years ago by sshleifer
  0