microsoft/ProphetNet

Issues

• AR-Diffusion Cannot Inference with the pretrained Commongen checkpoint

  #86 opened 4 months ago by chuanky
  0

• AI Augmented Cloud Administration Of Justice

  #85 opened 4 months ago by RonWiseAI
  0

• Unable to download ProphetNet checkpoints for pre-trained models

  #83 opened 6 months ago by alt-control
  0

• Unable to load the GENIE model

  #64 opened a year ago by CSerxy
  2

• what is the need for `num_samples` parameter in inference?

  #79 opened 9 months ago by hwaseem04
  1

• (AR-Diffusion) RuntimeError: Error(s) in loading state_dict for CrossAttention_Diffusion_LM

  #78 opened 9 months ago by hwaseem04
  3

• AR-diffusion: where the code for algorithm 1 is located?

  #77 opened 9 months ago by miaoGao
  4

• The datasets have no dev set？

  #65 opened a year ago by lcc0504
  1

• AR-Diffusion data.name and exp.name

  #75 opened 10 months ago by AIgods
  2

• [AR-Diffusion] predict_xstart vs predict_x_start

  #74 opened a year ago by raymond-yuan
  3

• “load_fairseq” not found in "AR-Diffusion/data_utils"

  #71 opened a year ago by Don-Joey
  3

• Request the execution code of llama2

  #76 opened 10 months ago by AnonymousLYZYY
  0

• “load_fairseq” not found in "AR-Diffusion/data_utils"

  #73 opened a year ago by ecust-sl
  1

• Unable to run Genie_Finetune.py

  #70 opened a year ago by xiang-xiang-zhu
  1

• Question for GENIE Finetuning, how to specify epochs for training/finetuning?

  #72 opened a year ago by SpaceHunterInf
  1

• Missing key documents for AR-Diffusion

  #68 opened a year ago by BiinRw
  1

• where is "mbr_select.py" in AR-Diffusion

  #69 opened a year ago by Jeongwan-Shin
  1

• It seems that the core code of CRITIC, particularly the part involving Google API search, is not implemented

  #67 opened a year ago by zyg18181818
  4

• Options Employed for Training or Inference on the CNN/DM Dataset

  #66 opened a year ago by Jeongwan-Shin
  1

• Where to obtain the test.tgt for CNN/DM data

  #56 opened 2 years ago by BaohaoLiao
  5

• Can't Find Pretrained Checkpoint of Prophetnet:

  #63 opened a year ago by hungptit123
  1

• Character level

  #62 opened a year ago by striver123456
  0

• Why is the GENIE result in AR-diffusion very different from the original paper? Also, you come from a team.

  #61 opened 2 years ago by zhaijianyang
  1

• Can use_fp16 be used?

  #59 opened 2 years ago by zhaijianyang
  0

• GENIE: decoder_nll loss is always equal to 0 for training from scratch

  #58 opened 2 years ago by BaohaoLiao
  2

• The reported scores of GENIE are not fair

  #57 opened 2 years ago by BaohaoLiao
  1

• KeyError during inference with dialog-en model

  #49 opened 3 years ago by raviteja5
  1

• Train new model

  #46 opened 3 years ago by stoicity
  1

• How to pretrain the ProphetNet?

  #54 opened 3 years ago by thomas-li-sjtu
  0

• Selecting additional scoring methods for fine-tuning

  #53 opened 3 years ago by mtsourma
  0

• How to use ProphetNetX in other languages like Vietnamese? How to preprocess the data?

  #45 opened 3 years ago by stoicity
  0

• About ProphetNet-Dialog-En in dialogue dataset

  #51 opened 3 years ago by xiang-xiang-zhu
  0

• Wrong Tokenization in SquadQG Evaluation Scripts

  #50 opened 3 years ago by hzhwcmhf
  0

• 1. Post a typo. 2. Wander the time cost. 3. Error in inference

  #48 opened 3 years ago by PolarisRisingWar
  2

• How to prepare and train with other language raw datasets?

  #32 opened 4 years ago by stoicity
  1

• Where can I get the evaluation script for English Dialog, including the AVG？

  #47 opened 3 years ago by FlyingCat-fa
  0

• KeyError: 'best_loss' at inference while using prophetnet code pretrained checkpoints.

  #44 opened 3 years ago by anks12297
  3

• Official implementation of Huggingface version

  #42 opened 3 years ago by Doragd
  0

• How to use the pretraining task of ProphetNet

  #43 opened 3 years ago by StevenTang1998
  0

• Is there any instructions to create models for text summarization in other languages (Vietnamese for example)?

  #41 opened 3 years ago by stoicity
  5

• Train model for Vietnamese summarization

  #36 opened 4 years ago by stoicity
  3

• About ProphetNet_Code's README

  #40 opened 4 years ago by ppaulggit
  1

• Inference AttributeError: 'Namespace' object has no attribute 'max_source_positions'

  #39 opened 4 years ago by LuRuz
  0

• Variables needed for gradient computation should be modified by an inplace operation.

  #38 opened 4 years ago by aleversn
  0

• Where can I get the evaluation script for Gigaword dataset?

  #29 opened 4 years ago by stoicity
  2

• How I can build models for other languages?

  #34 opened 4 years ago by stoicity
  0

• Where is the embedding process? Which embedding method used by ProphetNet?

  #35 opened 4 years ago by stoicity
  0

• chinese

  #33 opened 4 years ago by Wang-Yufei
  0

• Assertion Error in fine-tuning of Gigaword

  #31 opened 4 years ago by takase
  2

• About time cost on pre-training with corpus Wikipedia.

  #30 opened 4 years ago by bescks
  0