Ecosyste.ms: Issues

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

GitHub / microsoft/DeBERTa issues and pull requests

#157 - Runner Wrapper to call run.py

Pull Request - State: closed - Opened by zkalson 11 days ago

#156 - Update requirements.txt - sklearn to scikit-learn

Pull Request - State: open - Opened by souvikb1812 3 months ago

#155 - How to pre training mDeBERTa model?

Issue - State: open - Opened by 9mean2 4 months ago - 1 comment

#154 - RTD is not registed

Issue - State: closed - Opened by Eric-Chen-007 5 months ago - 1 comment

#153 - Pretraining the deberta-v3 by larger context length.

Issue - State: open - Opened by sherlcok314159 7 months ago - 2 comments

#152 - Generator weights

Issue - State: open - Opened by ir2718 10 months ago

#151 - Fine-tune DeBERTa v3 language model, worthwhile endeavour?

Issue - State: open - Opened by shensmobile 12 months ago - 7 comments

#150 - How can I evaluate COPA dataset?

Issue - State: open - Opened by KwanghyeonLee about 1 year ago

#145 - Question regarding symmetric KL Loss

Issue - State: open - Opened by skbaur about 1 year ago

#144 - sklearn changed to scikit-learn to avoid pip install failure

Pull Request - State: open - Opened by RajkumarGalaxy over 1 year ago

#143 - Fix typo

Pull Request - State: open - Opened by mrm8488 over 1 year ago

#142 - fix: Corrected code instructions in README.md

Pull Request - State: open - Opened by zzc0430 over 1 year ago

#141 - fix: sklearn -> scikit-learn

Pull Request - State: open - Opened by zzc0430 over 1 year ago

#140 - Trying to initialize model "large"

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

#139 - EOF error while running the rtd.sh script

Issue - State: open - Opened by BartWesthoff over 1 year ago - 1 comment

#138 - change: reduce memory consumption

Pull Request - State: open - Opened by YongWookHa over 1 year ago - 2 comments

#137 - Trying to run rtd_task.py on Windows

Issue - State: open - Opened by Yuri-Albuquerque over 1 year ago - 1 comment

#135 - Eligibility for Commercial Use

Issue - State: closed - Opened by Hegelim almost 2 years ago - 1 comment

#134 - Install fails due to use of deprecated `sklearn` package

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

#133 - Change from deprecated package name sklearn to correct scikit-learn

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

#132 - Load deberta-v3-large but got deberta-v2 model

Issue - State: open - Opened by ChengsongLu almost 2 years ago - 2 comments

#131 - Deberta-v3-base Generator model

Issue - State: open - Opened by sharanyarc96 almost 2 years ago - 2 comments

#130 - n/a

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

#129 - AssertionError: RTD is not registed.

Issue - State: closed - Opened by StephennFernandes almost 2 years ago - 3 comments

#128 - No module named 'torch._six'

Issue - State: closed - Opened by StephennFernandes almost 2 years ago - 3 comments

#127 - Error when running the example code for pretraining the rtd model.

Issue - State: open - Opened by soonilbae almost 2 years ago - 15 comments

#126 - effectiveness of RTD

Issue - State: open - Opened by martin-reczko almost 2 years ago

#125 - Info on Deberta-v2-xlarge training infra

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

#124 - Microsoft

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

#123 - mDeBERTa Generator model

Issue - State: closed - Opened by dadelani almost 2 years ago - 3 comments

#122 - 1. Add code for DeBERTaV3 pre-training; 2. Fix error in torch 1.11; 3…

Pull Request - State: closed - Opened by BigBird01 almost 2 years ago

#121 - Generator Model

Issue - State: closed - Opened by prajwal967 about 2 years ago - 1 comment

#120 - Convert DeBERTa model to ONNX with mixed precision

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

#119 - which version is torch ?

Issue - State: closed - Opened by XuJianzhi about 2 years ago

#118 - Fix: a few typos as I read through the README.md

Pull Request - State: closed - Opened by cpcdoy about 2 years ago

#117 - why vocab.txt and tokenizer.json not in pretrained model in huggingface ??

Issue - State: open - Opened by XuJianzhi about 2 years ago - 1 comment

#116 - Code about deberta_v3

Issue - State: closed - Opened by BAOOOOOM about 2 years ago - 1 comment

#115 - AssertionError: [] in google coab

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

#113 - mDeBERTa large

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

#111 - Where is the Gradient-Disentangled Embedding Sharing(GDES) part in the code?

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

#110 - Can't run bash commands in /DeBERTa/experiments/glue/

Issue - State: closed - Opened by heya5 over 2 years ago

#109 - out of memory

Issue - State: open - Opened by Amazing-J over 2 years ago - 18 comments

#108 - How to pretrain DeBERTa v3 ??

Issue - State: closed - Opened by BinhMinhs10 over 2 years ago - 2 comments

#106 - Why does the size of DeBERTaV3 double on disk after finetuning?

Issue - State: closed - Opened by nadahlberg almost 3 years ago - 2 comments

#105 - where is ENHANCED MASK DECODER ACCOUNTS part in code?

Issue - State: closed - Opened by tjshu almost 3 years ago - 1 comment

#104 - Evaluation hangs for distributed MLM task

Issue - State: open - Opened by dannyel2511 almost 3 years ago - 7 comments

#103 - Embedding layer vocab size not match to tokenizer length

Issue - State: open - Opened by kingbone9 almost 3 years ago - 1 comment

#102 - DeXLNeta

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

#100 - Pre-training times: v2 vs. v3

Issue - State: open - Opened by stefan-it almost 3 years ago - 1 comment

#99 - This model for MLM is waste of time, why did you even made it if it cannot be used?

Issue - State: closed - Opened by Oxi84 almost 3 years ago - 6 comments

#98 - How to use this model for MLM task?

Issue - State: open - Opened by Oxi84 almost 3 years ago

#97 - Release source distribution through PyPI or GitHub releases

Issue - State: open - Opened by BastianZim almost 3 years ago - 1 comment

#96 - Pretrained RTD Model ?

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

#95 - Import ABC from collections.abc for Python 3.10 compatibility

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

#94 - DeBERTaV3 small & xsmall pre-training configuration?

Issue - State: open - Opened by zealotjin almost 3 years ago - 2 comments

#93 - How to pretrain mDeBERTa ?

Issue - State: open - Opened by StephennFernandes almost 3 years ago - 24 comments

#92 - DebertaV3: is EMD used in generator and discriminator?

Issue - State: open - Opened by stevenliu000 almost 3 years ago

#91 - Machine translation

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

#90 - checkpoint in where?

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

#89 - Sharing DeBERTa-v3 discriminator and generator with task-specific heads?

Issue - State: closed - Opened by MoritzLaurer about 3 years ago - 10 comments

#88 - how to solve this bug

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

#87 - Add fine-tuning script for alpha nli

Pull Request - State: closed - Opened by BigBird01 about 3 years ago

#86 - Cannot get the [MASK] token correctly tokenized

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

#85 - Why is there an extra layernorm in sift/PerturbationLayer?

Issue - State: open - Opened by s-JoL about 3 years ago

#82 - SPM Tokenizer

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

#81 - Index tensor must have the same number of dimensions as self tensor

Issue - State: closed - Opened by Bushra178 about 3 years ago - 1 comment

#80 - Entailment Scores different in CPU and GPU device

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

#79 - Need DeBERTa v3 xsmall Huggingface Transformers fastTokenizer for NER

Issue - State: closed - Opened by bluesky333 about 3 years ago - 1 comment

#78 - Add script for XSmall model fine-tuning. Optimize the code for better…

Pull Request - State: closed - Opened by BigBird01 about 3 years ago

#77 - mDeBERTa on HuggingFace hub does not seem to work

Issue - State: closed - Opened by MoritzLaurer about 3 years ago - 18 comments

#76 - Add document for DeBERTa pre-training

Pull Request - State: closed - Opened by BigBird01 about 3 years ago

#75 - MLM Pre-training Code Version

Issue - State: closed - Opened by robinsongh381 about 3 years ago - 3 comments

#74 - Pretrained models for masked LM do not work as expected

Issue - State: open - Opened by AnandA777 about 3 years ago - 9 comments

#73 - Fix broken link for mDeBERTa-V3-Base in README

Pull Request - State: closed - Opened by vilhub about 3 years ago

#72 - the results of debertav3 small on the mnli task

Issue - State: closed - Opened by nbcc about 3 years ago - 2 comments

#71 - DeBERTa-V3 Pre-training code ?

Issue - State: open - Opened by kamalkraj about 3 years ago - 16 comments

#70 - Can't load DeBERTa-v3 tokenizer

Issue - State: closed - Opened by maiiabocharova about 3 years ago - 4 comments

#69 - DeBERTa V3 Fine-Tuning

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

#68 - Penhe/fix lm training

Pull Request - State: closed - Opened by BigBird01 over 3 years ago

#67 - DeBERTa RTD

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

#66 - Update setup.py

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

#65 - how to run the GLUE diagnostics dataset?

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

#64 - How to evaluate squad

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

#63 - How to train

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

#62 - assertion hit during forward pass

Issue - State: closed - Opened by xzhu1900 over 3 years ago - 1 comment

#61 - DeBERTa-MT and code for NLG experiments

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

#60 - How to get sentence embedding from DeBERTa for SNLI Dataset?

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

#59 - Why mlm task can't be registed

Issue - State: open - Opened by VV0808 over 3 years ago - 1 comment

#58 - how to train wsc in superglue

Issue - State: open - Opened by xiongjun19 over 3 years ago

#57 - Questions about pretraining dataset preparation

Issue - State: open - Opened by mansimane almost 4 years ago