Commits · 1eb91d3657fff606083a1d2d2b6da515f2cf3598 · gaoqiong / lm-evaluation-harness

12 Feb, 2021 6 commits
- adds missing task error · 1eb91d36
  & authored Feb 12, 2021
  
  1eb91d36
- Fixed metrics. Translation works · 61aa411b
  & authored Feb 12, 2021
  
  61aa411b
- add tasks to registry · 1297e342
  & authored Feb 12, 2021
  
  1297e342
- metrics · 12ba8426
  & authored Feb 12, 2021
  
  12ba8426
- translation task draft 1. Not hooked up, no metrics · bd3fcce0
  & authored Feb 12, 2021
  
  bd3fcce0
- metrics file · 1fb90b91
  & authored Feb 12, 2021
  
  1fb90b91
11 Feb, 2021 5 commits
- Merge branch 'master' of github.com:EleutherAI/lm_evaluation_harness · f7992789
  Leo Gao authored Feb 10, 2021
  
  f7992789
- Fixes to make greedy_until work · 432bd44c
  Leo Gao authored Feb 10, 2021
```
# Conflicts:
#	lm_eval/models/gpt2.py
#	lm_eval/tasks/squad.py
```
  432bd44c
- Merge pull request #142 from jon-tow/hellaswag-refactor · f63bc658
  Leo Gao authored Feb 10, 2021
```
Refactor `HellaSwag` as a `MultipleChoiceTask`
```
  f63bc658
- Implement GPT2 greedy_until · e8f9dc71
  Leo Gao authored Feb 10, 2021
  
  e8f9dc71
- Remove test set from `HellaSwag` · a2b108b9
  Jon Tow authored Feb 10, 2021
  
  a2b108b9
10 Feb, 2021 18 commits
- Refactor `HellaSwag` as a `MultipleChoiceTask` · 43ffe351
  Jon Tow authored Feb 10, 2021
  
  43ffe351
- Merge pull request #141 from jon-tow/winograd-fixes2 · 9adf18b1
  Leo Gao authored Feb 10, 2021
```
Fix context orderings in `Winograds` to match label positioning
```
  9adf18b1
- Rename list of loglikelihoods for consistency · 155aeee9
  Jon Tow authored Feb 10, 2021
  
  155aeee9
- Fix context orderings to match label positioning · 43e274f8
  Jon Tow authored Feb 10, 2021
  
  43e274f8
- Merge pull request #138 from jon-tow/winograd-fixes · c6b094e0
  Leo Gao authored Feb 09, 2021
```
Fix few-shot contexts for `Winograd Schema Challenge`-based tasks
```
  c6b094e0
- Merge branch 'master' of github.com:EleutherAI/lm_evaluation_harness into winograd-fixes · 1a159d6b
  Leo Gao authored Feb 09, 2021
  
  1a159d6b
- Update spacing · 8b038c2a
  Leo Gao authored Feb 09, 2021
  
  8b038c2a
- Remove incorrect context from `fewshot_context` · 3a90e246
  Jonathan Tow authored Feb 10, 2021
  
  3a90e246
- Merge pull request #139 from jeffhsu3/pubmedqa · 7614a8f3
  Leo Gao authored Feb 09, 2021
```
QA4MRE Task
```
  7614a8f3
- Update qa4mre.py · 5960874b
  Leo Gao authored Feb 09, 2021
  
  5960874b
- Add missing space in MultiRC · e9e5d0a2
  Leo Gao authored Feb 09, 2021
  
  e9e5d0a2
- Fix arithmetic · 42df113c
  Leo Gao authored Feb 09, 2021
  
  42df113c
- Cache by default · 84f14daf
  Leo Gao authored Feb 09, 2021
  
  84f14daf
- Update gpt2 for efficiency and allow specifying model size · aab91285
  Leo Gao authored Feb 09, 2021
  
  aab91285
- removed truncation of source · a264f449
  jeffhsu3 authored Feb 10, 2021
  
  a264f449
- Fix few-shot contexts for `Winograd Schema Challenge`-based tasks · 74f3307b
  Jonathan Tow authored Feb 09, 2021
  
  74f3307b
- Fix PiQA · 4d8ed7d5
  Leo Gao authored Feb 09, 2021
  
  4d8ed7d5
- Update RACE format · 62714855
  Leo Gao authored Feb 09, 2021
  
  62714855
09 Feb, 2021 7 commits
- space in pubmedqa for merge · 18e08d29
  jeffhsu3 authored Feb 09, 2021
  
  18e08d29
- Added qa4mre and changed to best_download for sciq · c05b9940
  jeffhsu3 authored Feb 09, 2021
  
  c05b9940
- More task changes · 9b14235d
  Leo Gao authored Feb 08, 2021
  
  9b14235d
- Fix · 96157fc7
  Leo Gao authored Feb 08, 2021
  
  96157fc7
- Change glue and superglue prompts · c3f724cf
  Leo Gao authored Feb 08, 2021
  
  c3f724cf
- Merge pull request #136 from jon-tow/openbookqa-evaluation · 1050109b
  Leo Gao authored Feb 08, 2021
```
Implement `OpenBookQA` evaluation
```
  1050109b
- Implement `OpenBookQA` evaluation · aa1d7293
  Jon Tow authored Feb 08, 2021
  
  aa1d7293
08 Feb, 2021 4 commits
- LM: handle empty context · 359114fd
  Leo Gao authored Feb 07, 2021
  
  359114fd
- Add table making · 5d56a47d
  Leo Gao authored Feb 07, 2021
  
  5d56a47d
- LAMBADA metric consistency change · dcd44d22
  Leo Gao authored Feb 07, 2021
  
  dcd44d22
- Add tests for {val,test}_docs determinism · 442f843c
  Leo Gao authored Feb 07, 2021
  
  442f843c