princeton-nlp/lm-bff issues and pull requests

#54 - generate_k_shot_data.py

Issue - State: open - Opened by why521-may about 2 months ago - 3 comments

#53 - Adapting LM-BFF to WiC dataset

Issue - State: open - Opened by schramm12 12 months ago

#52 - cannot import name 'default_hp_search_backend' from 'transformers.integrations'

Issue - State: open - Opened by schramm12 12 months ago - 4 comments

#51 - Problem with the requirements file

Issue - State: open - Opened by schramm12 12 months ago - 1 comment

#50 - Some questions regarding details of your paper results

Issue - State: closed - Opened by shijieli94 about 1 year ago - 2 comments

#49 - How can I create templates for right to left languages?

Issue - State: open - Opened by moeyad7 about 1 year ago

#48 - Hyperparameter selection for Finetuning Full setting.

Issue - State: closed - Opened by sh0416 almost 2 years ago - 1 comment

#47 - Weird SST2 dataset size

Issue - State: closed - Opened by sh0416 almost 2 years ago - 4 comments

#46 - question about label words which consist of more than 2 sub tokens

Issue - State: closed - Opened by JH-lee95 about 2 years ago - 1 comment

#45 - The issue for the loss of regression tasks

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

#44 - Bug in get_sbert_embedding.py?

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

#43 - lm_bff.json文件中相互关联的参数："inspired_templates", "replace_token_map_list"如何设置？

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

#42 - A little question about the evaluation

Issue - State: closed - Opened by shadowkiller33 over 2 years ago - 1 comment

#41 - Bug in tools/ensemble.py/get_labels

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

#40 - num_k parameter is not used

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

#39 - label generation problem

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

#38 - Average the logits of 16 demonstrations

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

#37 - enable truncate_head param when using normal prompt (w/o demo)

Pull Request - State: closed - Opened by Raibows over 2 years ago - 2 comments

#36 - Table 3 uses manual label words or generated label words?

Issue - State: closed - Opened by wjczf123 almost 3 years ago - 2 comments

#35 - changes for newer package versions with CUDA support

Pull Request - State: open - Opened by JingchaoZhang almost 3 years ago - 3 comments

#34 - Bug in dataloader ?

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

#33 - Negative CoLA MCC - Prompt-Based Finetuning error?

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

#32 - Question about prompt-based finetuning and automatic selection of label words

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

#31 - What is difference between --type prompt & --type prompt-demo?

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

#30 - Using trained models in a separate process for inference

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

#29 - Error while calling dataset.py: ValueError: 50264 is not in list

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

#28 - Testing: New Data for GLUE Tasks

Issue - State: closed - Opened by YashBit about 3 years ago - 6 comments

#27 - ImportError: cannot import name 'BertOnlyMLMHead' from 'transformers'

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

#26 - Can only use one label word mapping?

Issue - State: closed - Opened by hl0929 about 3 years ago - 1 comment

#25 - Question about the regression problem

Issue - State: closed - Opened by Elenore1997 about 3 years ago - 2 comments

#24 - ValueError: 50264 is not in list

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

#23 - 模板生成中的疑问

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

#22 - Question about the Ensemble results

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

#21 - Changes to make this repo work with CTA

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

#20 - 国内下载速度好慢，大家有下好的数据吗

Issue - State: closed - Opened by KyrieIrving24 over 3 years ago - 2 comments

#19 - ValueError Some specified arguments are not used by the HfArgumentParser

Issue - State: closed - Opened by demongolem-biz over 3 years ago - 3 comments

#18 - Create LICENSE

Pull Request - State: closed - Opened by gaotianyu1350 over 3 years ago

#17 - 一些细节的关于paper和代码的问题

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

#16 - Load and predict

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

#15 - Training sample template conversion

Issue - State: closed - Opened by v1nc3nt27 over 3 years ago - 2 comments

#15 - Training sample template conversion

Issue - State: closed - Opened by v1nc3nt27 over 3 years ago - 2 comments

#14 - [SST-2] Different test.tsv for original and k-shot data

Issue - State: closed - Opened by skull8888888 over 3 years ago - 3 comments

#13 - Is there a way to deal with label words with multiple tokens?

Issue - State: closed - Opened by ryangawei over 3 years ago - 6 comments

#12 - Specify Model Training Epochs

Issue - State: closed - Opened by ryangawei almost 4 years ago - 2 comments

#11 - Templates & Label Word Generation with Multilingual T5

Issue - State: closed - Opened by ryangawei almost 4 years ago - 2 comments

#10 - ValueError: 103 is not in list / 50264 not in list

Issue - State: closed - Opened by nelson-liu almost 4 years ago - 2 comments

#9 - Clarification re: --num_samples value

Issue - State: closed - Opened by nelson-liu almost 4 years ago - 2 comments

#8 - About the SNLI datasets

Issue - State: closed - Opened by MatthewCYM almost 4 years ago - 1 comment

#7 - Index not in list error when evaluating models zero-shot

Issue - State: closed - Opened by rtaori almost 4 years ago - 2 comments

#6 - Which pretrained models can be used with this codebase?

Issue - State: closed - Opened by rtaori almost 4 years ago - 2 comments

#5 - Bug in FewShotDataset module when accessing TextClassificationProcessor

Issue - State: closed - Opened by ramakanth-pasunuru almost 4 years ago - 3 comments

#4 - SNLI metrics are not supported in Transformer's glue_compute_metrics module

Issue - State: closed - Opened by ramakanth-pasunuru almost 4 years ago - 2 comments

GitHub / princeton-nlp/lm-bff issues and pull requests