Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / allenai/longformer issues and pull requests

#259 - Cosine similarity scores between random words are well above 0.9

Issue - State: open - Opened by diogo-p-nunes 5 months ago - 1 comment

#258 - "requirements.txt" update (transformers==3.0.2)

Issue - State: open - Opened by XinyiYing 11 months ago - 2 comments

#257 - Can Longformer support adapter transformer?

Issue - State: open - Opened by 14H034160212 over 1 year ago

#255 - Reproducibility Problem

Issue - State: open - Opened by wuj44 over 1 year ago

#254 - Where is the global attention?

Issue - State: open - Opened by Trace2333 over 1 year ago

#251 - On cheatsheet

Issue - State: open - Opened by chlorane almost 2 years ago

#250 - Mes/longformer on beaker copy all

Pull Request - State: open - Opened by ItzVladick almost 2 years ago

#249 - T5 encoder decoder

Pull Request - State: open - Opened by ItzVladick almost 2 years ago

#248 - Number of tokens per batch mismatch - longformer vs roberta

Issue - State: open - Opened by nbroad1881 almost 2 years ago - 1 comment

#247 - Answering performance of Longformer-base on the HotpotQA dev set

Issue - State: closed - Opened by zycdev almost 2 years ago

#246 - CVE-2007-4559 Patch

Pull Request - State: open - Opened by TrellixVulnTeam about 2 years ago

#245 - Updated BART to Longformer-encoder-decoder (LED) converter

Issue - State: open - Opened by erichans about 2 years ago

#244 - Why the TVM impelmentation is memroy efficient

Issue - State: open - Opened by jlidw about 2 years ago

#243 - Pretraining longformer for NER on big pdf text

Issue - State: open - Opened by ajaysurya1221 about 2 years ago

#242 - LED Training Time

Issue - State: open - Opened by gospelnnadi about 2 years ago

#241 - One hot encoding classes

Issue - State: open - Opened by PersianSpock about 2 years ago

#240 - Can't find a valid checkpoint at tmp

Issue - State: open - Opened by FarnazZeidi about 2 years ago

#238 - AttributeError: 'RobertaEmbeddings' object has no attribute 'position_ids'

Issue - State: closed - Opened by XueqiYang over 2 years ago - 1 comment

#236 - ERROR: Command errored out with exit status 128: git

Issue - State: open - Opened by CNwangbin over 2 years ago

#235 - ModuleNotFoundError: No module named 'longformer'

Issue - State: closed - Opened by CNwangbin over 2 years ago - 1 comment

#234 - Results on Hyperpartisian

Issue - State: open - Opened by amineabdaoui over 2 years ago - 3 comments

#232 - Where is the embedding?

Issue - State: closed - Opened by vqiangv over 2 years ago

#230 - where is the “transformers.modeling_roberta”？

Issue - State: closed - Opened by vqiangv over 2 years ago - 3 comments

#229 - Error loading data in summarization.py

Issue - State: closed - Opened by yjqiu over 2 years ago - 1 comment

#228 - Fine-tuning longformer for Question Answering

Issue - State: open - Opened by SumeetSandhu over 2 years ago

#226 - Where is the longformer-loop version?

Issue - State: open - Opened by diaodeyi over 2 years ago

#225 - Cuda 11 support for Ubuntu 18.04 and 20.04

Pull Request - State: closed - Opened by jannessm almost 3 years ago

#224 - Indeterministic results with LongFormer

Issue - State: open - Opened by Copilot-X almost 3 years ago

#223 - integrate with Lightning ecosystem CI

Issue - State: open - Opened by Borda almost 3 years ago

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

Issue - State: open - Opened by LadyHangaku almost 3 years ago

#221 - Trying to run longformers on TPU

Issue - State: open - Opened by siddagra almost 3 years ago

#220 - Adapting this repo to current version of transformers library

Issue - State: open - Opened by MorenoLaQuatra almost 3 years ago - 3 comments

#219 - Autoregressive and Relative Position Embedding Support

Issue - State: open - Opened by btyu almost 3 years ago

#216 - LED models give: `IndexError: index out of range in self`

Issue - State: closed - Opened by nicola-decao about 3 years ago - 1 comment

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

Issue - State: open - Opened by SCS2017 about 3 years ago - 3 comments

#214 - is here a simple example for seq2seq?

Issue - State: open - Opened by seyeeet about 3 years ago - 1 comment

#213 - Longformer issue in Huggingface implementation

Issue - State: closed - Opened by aleSuglia about 3 years ago - 1 comment

#212 - Longformer for autoregression

Issue - State: open - Opened by siddagra about 3 years ago

#211 - Initialization for large-model-training is far too slow

Issue - State: open - Opened by CaoYiqingT about 3 years ago

#210 - Difference between this codebase and Huggingface?

Issue - State: open - Opened by aleSuglia about 3 years ago

#207 - LongformerForSequenceClassification explanation

Issue - State: open - Opened by Nick9214 over 3 years ago - 1 comment

#206 - Error when converting MBart to Longformer

Issue - State: open - Opened by edgartanaka over 3 years ago - 2 comments

#205 - Correct way of loading pretrained model led-base-16384

Issue - State: open - Opened by kgarg8 over 3 years ago

#203 - Embedding dimension

Issue - State: open - Opened by Nick9214 over 3 years ago

#200 - Update longformer.py

Pull Request - State: closed - Opened by Herais over 3 years ago

#199 - The availability of Longformer-tiny?

Issue - State: open - Opened by songhwanjun over 3 years ago

#198 - Reproducibility of Table 11 (Summarization)

Issue - State: open - Opened by fengwang99feng over 3 years ago

#197 - Gradio Web Demo

Pull Request - State: open - Opened by AK391 over 3 years ago

#196 - first commit

Pull Request - State: closed - Opened by aliebrahiiimi over 3 years ago

#195 - cannot import name 'nvcc' from 'tvm.contrib' (unknown location)

Issue - State: open - Opened by wsmzzz over 3 years ago - 4 comments

#193 - Run inference for summarization

Issue - State: open - Opened by jacob-parnell-rozetta over 3 years ago - 1 comment

#191 - Convert BERT to "long" version

Issue - State: open - Opened by dawei-yu over 3 years ago - 4 comments

#190 - CUDA out of memory with a paragraph of length 3000

Issue - State: closed - Opened by SefaZeng over 3 years ago - 2 comments

#189 - reproductivity of the output of Longformer

Issue - State: open - Opened by passenger20 over 3 years ago - 2 comments

#186 - Longformer model with weight(model.encoder.embed_positions.weight) error

Issue - State: open - Opened by BinchaoPeng over 3 years ago - 3 comments

#180 - local vs global attention in further MLM pre-training.

Issue - State: open - Opened by chrisvdwerf over 3 years ago - 3 comments

#179 - Long T5

Pull Request - State: open - Opened by HaokunLiu over 3 years ago

#177 - Size mismatch error - LongBART

Issue - State: open - Opened by amoramine over 3 years ago - 1 comment

#176 - Compile tvm kernel in newer version of CUDA

Issue - State: closed - Opened by elb3k over 3 years ago - 1 comment

#175 - longformer speed compared to bert model

Issue - State: open - Opened by gkim89 over 3 years ago - 1 comment

#171 - longformer infer speed?

Issue - State: open - Opened by lookmyeye over 3 years ago - 3 comments

#166 - Update conversion script to transformers v4.2.0

Pull Request - State: closed - Opened by adamwawrzynski almost 4 years ago - 1 comment

#163 - index out of range in self!

Issue - State: open - Opened by MarwaEssam almost 4 years ago - 6 comments

#157 - Question about the implemented sparse attention

Issue - State: open - Opened by lhl2017 almost 4 years ago - 5 comments

#152 - ModuleAttributeError: 'RobertaEmbeddings' object has no attribute 'position_ids'

Issue - State: open - Opened by yysirs almost 4 years ago - 10 comments

#150 - How to set attention mask, any suggestion?

Issue - State: open - Opened by thesby almost 4 years ago - 3 comments

#149 - T5

Pull Request - State: open - Opened by AkshitaB almost 4 years ago - 4 comments

#148 - Sentiment Analysis?

Issue - State: open - Opened by t-lochhead almost 4 years ago - 2 comments

#147 - Global attention in key_padding_mask

Issue - State: open - Opened by greenstars almost 4 years ago - 3 comments

#135 - Longformer is not converted to ONNX format.

Issue - State: open - Opened by vgavrilo about 4 years ago - 12 comments

#129 - the bug in convert_model_to_long.ipynb

Issue - State: open - Opened by hitskyer about 4 years ago - 3 comments

#114 - How to compare text similarity?

Issue - State: open - Opened by thesby about 4 years ago - 8 comments

#100 - `sliding_chunks_no_overlap` implementation of the local attention

Pull Request - State: closed - Opened by ibeltagy about 4 years ago

#98 - How to make LongAlbert?

Issue - State: open - Opened by talkhaldi about 4 years ago - 2 comments

#96 - Longformer Memory Consumption query

Issue - State: closed - Opened by PrudhviRaj12 about 4 years ago - 2 comments

#95 - Latest transformers convert error

Issue - State: open - Opened by Maybewuss over 4 years ago - 15 comments

#94 - Error with attention_mode in config.json from pretrained model

Issue - State: closed - Opened by Wilbur-Django over 4 years ago - 2 comments

#93 - Does Longformer predict the answer span on WikiHop dataset?

Issue - State: closed - Opened by sjy1203 over 4 years ago - 5 comments

#92 - Does transformers use the custom CUDA kernel?

Issue - State: closed - Opened by Maybewuss over 4 years ago - 5 comments

#91 - GPU OOM when training XLM-RoBERTa with LongSelfAttention

Issue - State: open - Opened by KasparPeterson over 4 years ago - 1 comment

#90 - Pretraining Dataset Details

Issue - State: closed - Opened by sjy1203 over 4 years ago - 3 comments

#89 - added output_attentions arg & super basic test

Pull Request - State: closed - Opened by riklopfer over 4 years ago - 1 comment