v0.10.6

Changes

• release 0.10.6 (#162) @ArneBinder
• Extend split_mappings format in concatenate_dataset_dicts() (#161) @RainbowRivey

🪲 Fixes

• sciarg: fix partitioning (#159) @ArneBinder
• upgrade GitHub actions (#160) @ArneBinder

👷 Continuous Integration

• upgrade GitHub actions (#160) @ArneBinder

v0.10.5

Changes

• release 0.10.5 (#157) @ArneBinder

🚀 Features

• Implement concatenate_dataset_dicts (#153) @RainbowRivey
• add parameter set_batch_size_to_split_size to DatasetDict.map (#155) @ArneBinder
• DatasetDict.to_json() can append to already serialized data (#156) @ArneBinder

v0.10.4

Changes

• release 0.10.4 (#151) @ArneBinder
• fix #149 drugprot polishment (#150) @kai-car

🎉 New Dataset

• Add CoMAGC dataset (#141) @kai-car

🚀 Features

• add concatenate datasets method (#148) @kai-car
• implement Dataset.from_documents (#136) @ArneBinder

v0.10.3

Changes

• release 0.10.3 (#146) @ArneBinder
• Add a test split loader for Drugprot (#142) @RainbowRivey
• Set specific pytorch version for major platforms (linux, win32, darwin) (#132) @kai-car

🎉 New Dataset

• Add Tbga dataset (#140) @kai-car
• Add Chemprot dataset (#138) @kai-car
• Add Biorel dataset (#134) @kai-car

🚀 Features

• add brat note (#143) @Bhuvanesh-Verma

🪲 Fixes

• poetry dependencies for torch in macOS with M-series (#144) @Bhuvanesh-Verma
• biorel: remove warning when entity name is not the same as the span text (#139) @ArneBinder

v0.10.2

Changes

• release 0.10.2 (#131) @ArneBinder

🪲 Fixes

• set max numpy version to <2.0.0 (#130) @ArneBinder

v0.10.1

Changes

• release 0.10.1 (#129) @ArneBinder
• update dependencies (#128) @ArneBinder

🎉 New Dataset

• add conll2012_ontonotesv5 dataset (#52) @ArneBinder

🪲 Fixes

• fix aae2 when passing conversion_method to load_dataset (#127) @ArneBinder

👷 Continuous Integration

• increase timeout for test_dataset CI job (#126) @ArneBinder
• fix datset test CI when no dataset was modified (#125) @ArneBinder
• test workflow per dataset (#122) @ArneBinder

v0.10.0

Changes

• release 0.10.0 (#124) @ArneBinder

🎉 New Dataset

• add DrugProt dataset (#106) @RainbowRivey

💥 Breaking Changes

• brat: use annotation types from pytorch-ie (#123) @ArneBinder

🚀 Features

• use local data for cdcp pie dataset tests (#115) @ArneBinder

🪲 Fixes

• fix CI for dataset tests (#118) @ArneBinder
• fix DatasetDict.select with document converters (#116) @ArneBinder

🚨 Testing

• use local data for cdcp pie dataset tests (#115) @ArneBinder

👷 Continuous Integration

• fix CI for dataset tests (#118) @ArneBinder
• separate test workflow for datasets (#117) @ArneBinder

v0.9.0

Changes

• release 0.9.0 (#114) @ArneBinder
• upgrade pie-modules to >=0.10.8,<0.12.0 (#113) @ArneBinder

💥 Breaking Changes

• update pytorch-ie to >=0.29.4,<0.31.0 and pie-modules to ^0.10.6 (#105) @ArneBinder

🚀 Features

• multi span handling for sciarg (#103) @ArneBinder

🐎 Performance

• decrease imdb test time (#109) @ArneBinder
• decrease squad_v2 test time (#108) @ArneBinder

🚨 Testing

• increase timeout for test CI job (#111) @ArneBinder
• decrease imdb test time (#109) @ArneBinder
• decrease squad_v2 test time (#108) @ArneBinder

v0.8.2

Changes

• release 0.8.2 (#104) @ArneBinder
• Add visualization for AM dataset cards. (#97) @idalr
• add aae2 dataset - v.2 (#92) @idalr

🚀 Features

• implement DatasetDict.move_shard_to_new_split() (#102) @ArneBinder

🪲 Fixes

• cdcp dataset: use fixed HF dataset (#101) @ArneBinder

v0.8.1

Changes

• release 0.8.1 (#95) @ArneBinder

🎉 New Dataset

• add abstrct dataset (#68) @ArneBinder
• Add scientific_papers dataset (#90) @jk-tripathy
• add sciarg dataset (#61) @ArneBinder
• add imdb dataset (#53) @ArneBinder
• add squad_v2 datasets (#51) @ArneBinder

🪲 Fixes

• pin datasets to <0.16.0 (#94) @ArneBinder
• fix sciarg tests: tokenization with partitions (#89) @ArneBinder