NielsRogge/transformers

Pull Requests Commits

Squash all commits of modeling_detr_v5 into one

NielsRogge committed 4 years ago

786c7058

Commit the last step on world_process_zero in WandbCallback (#9805)

tristandeleu committed 4 years ago

Verified eba418ac

Allow RAG to output decoder cross-attentions (#9789)

dblakely committed 4 years ago

Verified 8edc98bb

Fix fine-tuning translation scripts (#9809)

mbiesialska committed 4 years ago

Verified 8f6c12d3

Fixed parameter name for logits_processor (#9790)

michaelrglass committed 4 years ago

Verified c37dcff7

Smdistributed trainer (#9798)

sgugger committed 4 years ago

Verified 0d0efd3a

Fix head_mask for model templates

LysandreJik committed 4 years ago

897a24c8

Improve pytorch examples for fp16 (#9796)

ak314 committed 4 years ago

Verified 10e5f282

Adding `skip_special_tokens=True` to FillMaskPipeline (#9783)

Narsil committed 4 years ago

Verified 781e4b13

Add head_mask/decoder_head_mask for TF BART models (#9639)

stancld committed 4 years ago

Verified 1867d9a8

Fix broken links in the converting tf ckpt document (#9791)

forest1988 committed 4 years ago

Verified cb73ab5a

[Flaky Generation Tests] Make sure that no early stopping is happening for beam search (#9794)

patrickvonplaten committed 4 years ago

Verified d94cc2f9

[PR/Issue templates] normalize, group, sort + add myself for deepspeed (#9706)

stas00 committed 4 years ago

Verified 0fdbf085

sgugger committed 4 years ago

af41da50

Auto-resume training from checkpoint (#9776)

sgugger committed 4 years ago

Verified caf4abf7

Actual fix (#9787)

LysandreJik committed 4 years ago

Verified 0f443436

[fsmt] onnx triu workaround (#9738)

stas00 committed 4 years ago

Verified fac7cfb1

Fix a typo in Trainer.hyperparameter_search docstring (#9762)

sorami committed 4 years ago

Verified 626116b7

Use object store to pass trainer object to Ray Tune (#9749)

krfricke committed 4 years ago

Verified d63ab615

Fix TFTrainer prediction output (#9662)

janinaj committed 4 years ago

Verified 6312fed4

Fix broken [Open in Colab] links (#9761)

wilcoln committed 4 years ago

Verified 9152f160

token_type_ids isn't used (#9736)

stas00 committed 4 years ago

Verified b7b7e5d0

Fix test (#9755)

jplu committed 4 years ago

Verified a449ffcb

Add `report_to` training arguments to control the reporting integrations used (#9735)

sgugger committed 4 years ago

Verified 82d46feb

Fixes to run_seq2seq and instructions (#9734)

sgugger committed 4 years ago

Verified 411c5821

Fix some TF slow tests (#9728)

jplu committed 4 years ago

Verified d7c31abf

examples: fix XNLI url (#9741)

stefan-it committed 4 years ago

Verified 08b22722

Fix memory regression in Seq2Seq example (#9713)

sgugger committed 4 years ago

Verified 5f80c15e

Fix TF s2s models (#9478)

jplu committed 4 years ago

Verified a7dabfb3

Changing model default for TableQuestionAnsweringPipeline. (#9729)

Narsil committed 4 years ago

Verified 23e5a36e

Older