nyu-mll jiant-v1-legacy issues

nyu-mll / jiant-v1-legacy

The jiant toolkit for general-purpose text understanding models

MIT License

21 stars 9 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

scripts/download_glue_data.py

#1120 Xnhyacinth closed 1 year ago
0
EP preprocessing, cannot reproduce

#1119 terarachang closed 3 years ago
3
Update README.md

#1118 jeswan closed 3 years ago
0
Add a forward link in the readme to v2.0

#1117 sleepinyourhat closed 3 years ago
0
Mark "[CLOSED]" PRs as closed

#1116 sleepinyourhat closed 4 years ago
0
Error in edgeprobe_bert

#1115 jeswan opened 4 years ago
4
Is squad task implemented?

#1114 jeswan opened 4 years ago
1
How to get the JSON files for SPR1 and 2?

#1113 jeswan opened 4 years ago
1
[CLOSED] [fix] Update the glue data script to use new download links

#1112 jeswan closed 4 years ago
4
acquiring data on the lengths of spans

#1111 jeswan opened 4 years ago
1
[CLOSED] Reproducing GLUE paper single task results

#1110 jeswan closed 4 years ago
1
[CLOSED] Replicating RobertaForSequenceClassification model

#1109 jeswan closed 4 years ago
1
[CLOSED] problem extracting the POS data from the ontonotes corpus

#1108 jeswan closed 4 years ago
2
[CLOSED] Clarify setup and environment variable instructions

#1107 jeswan closed 4 years ago
0
[CLOSED] tutorial documentation

#1106 jeswan closed 4 years ago
7
ImportError: cannot import name 'Ontonotes' from 'allennlp.data.dataset_readers.dataset_utils'

#1105 jeswan opened 4 years ago
2
[CLOSED] WiC and winogender-diagnostic - TypeError: apply_standard_boundary_tokens() got an unexpected keyword argument 'get_offset'

#1104 jeswan closed 4 years ago
3
[CLOSED] Update probing readme

#1103 jeswan closed 4 years ago
0
ontonotes coreference data

#1102 jeswan opened 4 years ago
1
[CLOSED] Updates to support using MiniBERTas as input_module

#1101 jeswan closed 4 years ago
7
[CLOSED] Update __init__.py so that MiniBERTas can be probed

#1100 jeswan closed 4 years ago
6
Problem when speeding up fine-tuning bert-base-uncased on ReCoRD

#1099 jeswan opened 4 years ago
1
support languages other than English

#1098 jeswan opened 4 years ago
1
Add option to re-initialize top N layers in Transformers models

#1097 jeswan opened 4 years ago
0
Add a warning in the defaults.conf for BERTAdam, and remove it as a default from all scripts

#1096 jeswan opened 4 years ago
0
[CLOSED] Support for external models query

#1095 jeswan closed 4 years ago
1
[CLOSED] Patch 1

#1094 jeswan closed 4 years ago
1
training with a subset of the training data

#1093 jeswan opened 4 years ago
2
KeyError: 'inputs' - while using BERT as input module

#1092 jeswan closed 4 years ago
6
Fixing model checkpoints to be robust to mutli -> single GPU usage

#1091 jeswan opened 4 years ago
3
[CLOSED] spr&srl data

#1090 jeswan closed 4 years ago
2
[CLOSED] Pin sklearn version in setup.py for compatibility with allennlp==0.8.4

#1089 jeswan closed 4 years ago
0
[CLOSED] Tests failing for master at HEAD

#1088 jeswan closed 4 years ago
0
Incompatibility in model checkpoints between multi-GPU/single-GPU

#1087 jeswan opened 4 years ago
4
[CLOSED] Switch to jsonl format for SNLI

#1086 jeswan closed 4 years ago
4
[CLOSED] Parameter parsing error with custom config.

#1085 jeswan closed 4 years ago
1
[CLOSED] reproducing the expected layer results of the paper

#1084 jeswan closed 4 years ago
3
[CLOSED] Update load_tsv to stop if "bad lines" are found

#1083 jeswan closed 4 years ago
2
[CLOSED] load_tsv skips "bad lines"

#1082 jeswan closed 4 years ago
0
MultiRC task- v1.0 results are better than v1.3 (Unknown reason, after reviewing changes)

#1081 jeswan opened 4 years ago
1
[CLOSED] Suppress verbose INFO-level logging by AllenNLP components in trainer.py

#1080 jeswan closed 4 years ago
1
[CLOSED] Unable to Use XLM as an input_module

#1079 jeswan closed 4 years ago
3
[CLOSED] Update refs in setup tutorial to v1.3.2

#1078 jeswan closed 4 years ago
0
[CLOSED] Increasing sample size for MLM test

#1077 jeswan closed 4 years ago
5
[CLOSED] Transient test failures from MLM/#1030

#1076 jeswan closed 4 years ago
1
[CLOSED] Backronym "jiant"

#1075 jeswan closed 4 years ago
0
Remove --remote_log from exp_fns.sh

#1074 jeswan opened 4 years ago
2
Potentially corrupts SNLI data

#1073 jeswan opened 4 years ago
10
[CLOSED] getting the probed model

#1072 jeswan closed 4 years ago
2
[CLOSED] Fix number of training examples in target training

#1071 jeswan closed 4 years ago
0