allenai/longformer

Issues

• Cosine similarity scores between random words are well above 0.9

  #259 opened 7 months ago by diogo-p-nunes
  1

• "requirements.txt" update (transformers==3.0.2)

  #258 opened a year ago by XinyiYing
  2

• Unable to use longformer in contextualized topic models - CUDA error: device-side assert triggered

  #233 opened 3 years ago by siames3
  1

• Can Longformer support adapter transformer?

  #257 opened 2 years ago by 14H034160212
  0

• may you share link to somebody else latest development for long sentences pls?

  #256 opened 2 years ago by Sandy4321
  0

• where is the “transformers.modeling_roberta”？

  #230 opened 3 years ago by vqiangv
  3

• Reproducibility Problem

  #255 opened 2 years ago by wuj44
  0

• TypeError: forward() takes from 2 to 7 positional arguments but 8 were given

  #215 opened 3 years ago by SCS2017
  3

• Where is the global attention?

  #254 opened 2 years ago by Trace2333
  0

• Longformer embeddings for calculating similarity score between 2 documents using KNN

  #253 opened 2 years ago by swethakasireddi
  0

• @ibeltagy I have similar issues with converting the model to ONNX, I converted the model to ONNX model, but when I tried to infer with onnxruntime I got ScatterND error while session run. I am guessing there are some operations not supported by onnx.

  #252 opened 2 years ago by wmh02240
  0

• On cheatsheet

  #251 opened 2 years ago by chlorane
  0

• Answering performance of Longformer-base on the HotpotQA dev set

  #247 opened 2 years ago by zycdev
  0

• Number of tokens per batch mismatch - longformer vs roberta

  #248 opened 2 years ago by nbroad1881
  1

• Updated BART to Longformer-encoder-decoder (LED) converter

  #245 opened 2 years ago by erichans
  0

• Why the TVM impelmentation is memroy efficient

  #244 opened 2 years ago by jlidw
  0

• Pretraining longformer for NER on big pdf text

  #243 opened 2 years ago by ajaysurya1221
  0

• LED Training Time

  #242 opened 2 years ago by gospelnnadi
  0

• One hot encoding classes

  #241 opened 2 years ago by PersianSpock
  0

• Can't find a valid checkpoint at tmp

  #240 opened 2 years ago by FarnazZeidi
  0

• AttributeError: module 'dill._dill' has no attribute 'stack'

  #239 opened 2 years ago by yangjenhao
  0

• AttributeError: 'RobertaEmbeddings' object has no attribute 'position_ids'

  #238 opened 3 years ago by XueqiYang
  1

• CUDA error: device-side assert triggered in multi class text classification

  #237 opened 3 years ago by iteimouri
  0

• Results on Hyperpartisian

  #234 opened 3 years ago by amineabdaoui
  3

• ERROR: Command errored out with exit status 128: git

  #236 opened 3 years ago by CNwangbin
  0

• ModuleNotFoundError: No module named 'longformer'

  #235 opened 3 years ago by CNwangbin
  1

• Where is the embedding?

  #232 opened 3 years ago by vqiangv
  0

• The expanded size of the tensor (4096) must match the existing size (512) at non-singleton dimension 1.

  #231 opened 3 years ago by vqiangv
  0

• Error loading data in summarization.py

  #229 opened 3 years ago by yjqiu
  1

• Fine-tuning longformer for Question Answering

  #228 opened 3 years ago by SumeetSandhu
  0

• How to deal with 3-dimensional attention_mask in LongformerSelfAttention

  #227 opened 3 years ago by khang-nguyen2907
  0

• Adapting this repo to current version of transformers library

  #220 opened 3 years ago by MorenoLaQuatra
  3

• Where is the longformer-loop version?

  #226 opened 3 years ago by diaodeyi
  0

• Error when converting MBart to Longformer

  #206 opened 3 years ago by edgartanaka
  2

• Indeterministic results with LongFormer

  #224 opened 3 years ago by Copilot-X
  0

• integrate with Lightning ecosystem CI

  #223 opened 3 years ago by Borda
  0

• Self-made Longformer doesn’t take more than 512 token

  #222 opened 3 years ago by LadyHangaku
  0

• Trying to run longformers on TPU

  #221 opened 3 years ago by siddagra
  0

• Autoregressive and Relative Position Embedding Support

  #219 opened 3 years ago by btyu
  0

• Confusion about attention mask for pretraining LongformerForMaskedLM

  #218 opened 3 years ago by frederikkemarin
  0

• Is it possible to train LongformerEncoderDecoder starting from Japanese T5 checkpoint?

  #217 opened 3 years ago by SleepingSkipper
  0

• LED models give: `IndexError: index out of range in self`

  #216 opened 3 years ago by nicola-decao
  1

• is here a simple example for seq2seq?

  #214 opened 3 years ago by seyeeet
  1

• Longformer issue in Huggingface implementation

  #213 opened 3 years ago by aleSuglia
  1

• Longformer for autoregression

  #212 opened 3 years ago by siddagra
  0

• Initialization for large-model-training is far too slow

  #211 opened 3 years ago by CaoYiqingT
  0

• LongformerForSequenceClassification explanation

  #207 opened 3 years ago by Nick9214
  1

• Difference between this codebase and Huggingface?

  #210 opened 3 years ago by aleSuglia
  0

• Cannot build the docker image following the Cheatsheet.txt

  #209 opened 3 years ago by zheng-ningxin
  0

• What kind/format of text can replace your use of wikitext-103-raw-v1?

  #208 opened 3 years ago by jenka13all
  0