Skip to content

Releases: castorini/pygaggle

PyGaggle v0.0.2

30 Jan 19:20
fc03507
Compare
Choose a tag to compare

fc03507 - Release 0.0.2 (#152) (Ronak)
767e08d - Replicated PyGaggle: Baselines on MS MARCO Document Retrieval (#151) (Kai Sun)
46bb71d - MS MARCO Document Retrieval Replication (#137) (Dahlia Chehata)
5d23949 - PyGaggle: Neural Ranking Baselines on MS MARCO Passage Retrieval - Entire Dev Set (#148) (Kai Sun)
1414e32 - update tools, cleanup readme.md to include duo (#147) (Ronak)
ad37b67 - duo documentation, cleanup mono documentation (#146) (Ronak)
2dbd4c0 - Fixed Document Title Type Mismatch (#145) (Kai Sun)
f37bac7 - Replicated PyGaggle: Neural Ranking Baselines on MS MARCO Passage Retrieval - Dev Subset (#143) (Kai Sun)
62ec449 - update replication log (#142) (Dahlia Chehata)
c7fdc4f - update replication log (#136) (Dahlia Chehata)
fd4784d - update replication log (#141) (Dahlia Chehata)
623285a - CovidQA, duo cleanup (#140) (Ronak)
05a63f1 - fix filenames (#139) (Ronak)
968363e - refactor folder name (#134) (Ronak)
b0d8901 - refactor scifact 1 (#133) (Ronak)
a70269a - add verT5erini experiment (#131) (Xueguang Ma 马雪光)
072d889 - fix README (#130) (Ronak)
d20334a - Adding duo T5 (#127) (wiltan-uw)
3b1f69c - Replication of monoBERT and monoT5 Baselines for MSMARCO Passage Ranking Task (#128) (Rakeeb Hossain)
9a1fe70 - fix inconsistency re upgrading to transformer 4.0.0 (#124) (Xueguang Ma 马雪光)
aed21c0 - clean filelock output (#123) (Ronak)
0f738c0 - upgrade submodules (#122) (Ronak)
0341ed4 - update to transformers 4.0.0, modify environment.yml too (#120) (Ronak)
b235fae - update transformers==2.10.0 to transformers==4.0.0rc1 (#118) (Xueguang Ma 马雪光)
68b421e - Update README (#117) (Ronak)
bcbb07e - Update replication log and refactor instructions (#116) (Ray Yang)
7c30e1e - Refactored TPU documentation (#114) (Ronak)
d840b0c - Refactor monoT5 TPU instructions, clean up to Data Prep, add relevant data links (#113) (Ronak)
071d59f - pr (#111) (ESTella)
da0fb61 - add monot5 tpu train doc (#108) (Xueguang Ma 马雪光)
5e1e0dd - Update experiments-msmarco-passage-subset.md (#109) (Ray Yang)
19b16d2 - Add load commend to CC instruction (#101) (Qing Guo)
e75f633 - replication log (#105) (stephaniewhoo)
978a071 - Add constructor functions for model and tokenizer of MonoBERT/T5 (#93) (Yuxuan Ji)
6a487b7 - add the script to train d2q on msmarco (#103) (richard3983)
f552b5d - add doc&scripts for msmarco experiments on tpu (#100) (Xueguang Ma 马雪光)
a745d80 - Replication experiement results for CovidQA, MSMARCO document and MSMARCO subset (#102) (Jerry)
e815051 - Add instructions to replicate entire dev set on Compute Canada (#99) (Qing Guo)
3d4b7c0 - Update README.md (#98) (Rodrigo Frassetto Nogueira)
6638f6e - Add TREC Covid Reranking Task (#95) (Justin Borromeo)
f72153b - add doc of converting t5 to hg model (#96) (Xueguang Ma 马雪光)
73539ea - Fix default args in MonoBERT/T5 (#92) (Yuxuan Ji)
d88f8ce - fix typo in README (#91) (Ronak)
94270a2 - MSMARCO Passage/Document/CovidQA replications (#90) (wiltan-uw)
41513a9 - Simplify boilerplate for monoT5 and monoBERT (#83) (Yuxuan Ji)
a258c13 - add trec writer (#85) (Ronak)
4019c3f - replication results for MSMARCO document and CovidQA. (#84) (qguo96)
a1461f5 - Add replication logs for MSMARCO passage, document and CovidQA. (#82) (Lizzy Zhang)
8eeefa5 - CovidQA + MSMARCO doc Replication log (#78) (Yuxuan Ji)
cfb4b02 - Add to MSMarco-Doc/CovidQA Replication Log, update CovidQA BM25 (#77) (Justin Borromeo)
daeb78c - Add MS-MARCO passage replication (#79) (Yuxuan Ji)
360be7a - Removed Colab instructions (#81) (qguo96)
cc85405 - Replication for MS MARCO passage on Colab (#76) (qguo96)
94befbd - change raw to metadata (#75) (Ronak)
f692da6 - Fixed broken link in replication log: experiments-msmarco-passage.md (#74) (Jimmy Lin)
96a7e8d - Fix Pyserini compatibility issues (#71) (Justin Borromeo)
ae2dfc5 - Update experiments-msmarco-document.md (#69) (Jimmy Lin)
ed2f1e6 - Update README to fix bug (#68) (Negar Arabzadeh)
f7f8c49 - Improve reranking example in README (#65) (Jimmy Lin)
5b03294 - Bump up pyserini to 0.9.4.0 (#64) (Jimmy Lin)
377c283 - sync environment.yml to requirements.txt (#63) (Ronak)
bfce4af - Update Replication Log (#62) (mrkarezina)
54f609d - Update Replication Log (#61) (mrkarezina)
c1a54cb - Adds BERT reranker example (#59) (Rodrigo Frassetto Nogueira)
ed96740 - Create CovidQA Doc (#56) (HangCui0510)
4b8d67b - Adds reranker example (#58) (Rodrigo Frassetto Nogueira)
3e07b5c - Update replication log and requirements doc (#55) (HangCui0510)
f2e078e - Update experiments-msmarco-passage.md (#54) (Justin Borromeo)
70b2a9f - Fix tools 1 (#51) (Ronak)
3364e2f - switch to tools (#50) (Ronak)
f621265 - Add evaluate_document_ranker (#49) (Xueguang Ma 马雪光)
13e099b - Update eval, switch from ssh to https, modify instructions (#47) (Ronak)
f204e4b - Fix submodule name to be consistent with pyserini, minor changes to get it to work (#45) (Ronak)
4b332fe - Update .gitmodules (#44) (richard3983)
2f0fe60 - Revert "add submodule "anserini-eval" (#36)" (#43) (Ronak)
a6a59f6 - replicated results for MS MARCO neural passage ranking experiments (#41) (Kelvin Jiang)
82dc086 - change option from model-name-or-path to simpler model, fix flake8 len 120 (#40) (Ronak)
6c49f6c - Update MS MARCO Passage Replication Log (#37) (HangCui0510)
9b5eca6 - Update experiments-msmarco-passage.md (#38) (Ronak)
591e7ff - ignore logs from transformers (#32) (Xueguang Ma 马雪光)
f3485ac - Fix qa transformer (#34) (Ronak)
add59ba - Index dir option, change evaluate_passage_ranking to be consistent with CovidQA, update index to latest (#33) (Ronak)
d44c89c - Add mono models to huggingface model-zoo and incorporate into pipeline (#29) (Xueguang Ma 马雪光)
8dcfaa1 - clarify to flake8 max len 100 (#30) (Ronak)
2901832 - Update dataset to 0.2 (#23) (Ralph Tang)
4801daa - Add replication for MS MARCO passage re-ranking (#27) (richard3983)
6e9dfc6 - Add replication for MS MARCO passage re-ranking (#25) (Xueguang Ma 马雪光)
69de7db - Repo layout (adding logs, models, indexes, runs), MS MARCO passage replication doc (#24) (Ronak)
2905235 - Flake8 (#19) (Ronak)
58f0244 - add notebook for CovidQA (#18) (Ronak)
ef31ca1 - Writer (#17) (Ronak)
3469584 - fix zero div and t5 decoder (#16) (Ronak)
18f848a - Some readme edits (#13) (Nikhil Gupta)
55e4961 - MSMARCO support: monoBERT (#14) (Ronak)
34345c8 - Update README.md (Ralph Tang)
ea628f2 - Add model instructions to README.md (#10) (Ralph Tang)
f6d2168 - Fixes encoding error (#9) (Rodrigo Frassetto Nogueira)
6029693 - Update README.md (#7) (Ralph Tang)
708fa66 - Update README.md (#6) (Jimmy Lin)

PyGaggle v0.0.1

23 Apr 21:08
ae3a054
Compare
Choose a tag to compare
Change evaluation methodology (#2)

* Change evaluation methodology

- Make MRR, P@1, and R@3 default
- Add natural language queries and keyword queries
- Add 9 more QD pairs

* Physical sciences topics

* Add temperature questions

* Add more examples

* Implement QA transformer reranker

* Fix dataset typo

* Implement random baseline

* Implement more reranker model types

- Implement question answering reranker
- Implement sequence classification reranker

* Add final dataset examples

* Remove missing IDs from dataset

* Improve cosine similarity matrix provider name

* Add dataset statistics calculation

* Add version to dataset

* Add random MRR calculation

* Fix off-by-one in random MRR computation

* Prepare for release

- Add setuptools script
- Fix circular imports

* Add more detailed README

* Change license to Apache

* Fix classifier name

* Clarify README

Co-authored-by: Nikhil Gupta <[email protected]>
Co-authored-by: Edwin Zhang <[email protected]>