NielsRogge/transformers

Pull Requests Commits

LysandreJik committed 5 years ago

b3931505

LysandreJik committed 5 years ago

751ac4d5

Add integration tests for xlm roberta modelling and xlm roberta tokenzier (#3014)

patrickvonplaten committed 5 years ago

Verified c913eb9c

Change masking to direct labeling for TPU support. (#2982)

srush committed 5 years ago

Verified e8ce63ff

missing ner link (#2967)

autoih committed 5 years ago

Verified 7a7ee28c

Adding usage examples for common tasks (#2850)

LysandreJik committed 5 years ago

Verified 65e7c90a

[ci] Run slow tests every day

julien-c committed 5 years ago

e693cd1e

[ci] Attempt to fix #2844

julien-c committed 5 years ago

4fc63151

Test correct tokenizers after default switch (#3003)

LysandreJik committed 5 years ago

Verified b90745c5

False by default (#3002)

LysandreJik committed 5 years ago

Verified 3716c3d8

Release: v2.5.1

LysandreJik committed 5 years ago

f9ec5ca9

Fix for fast tokenizers save_pretrained compatibility with Python. (#2933)

mfuntowicz committed 5 years ago

Verified 4cd9c097

fix _update_memory fn call in transformer-xl (#2971)

andompesta committed 5 years ago

Verified ee60840e

add explaining example to XLNet LM modeling (#2997)

patrickvonplaten committed 5 years ago

Verified 6a50d501

Add preprocessing step for transfo-xl tokenization to avoid tokenizing words followed by punction to <unk> (#2987)

patrickvonplaten committed 5 years ago

Verified 65d74c49

Add local_files_only parameter to pretrained items (#2930)

Bram Vanroy committed 5 years ago

Verified a143d947

Create README.md

mrm8488 committed 5 years ago

286d1ec7

kwargs are passed to both model and configuration in AutoModels (#2998)

LysandreJik committed 5 years ago

Verified 7984a70e

Testing that batch_encode_plus is the same as encode_plus (#2973)

LysandreJik committed 5 years ago

Verified 21d8b6a3

Add slow generate tests for pretrained lm models (#2909)

patrickvonplaten committed 5 years ago

Verified 17c45c39

Warning on `add_special_tokens` (#2966)

LysandreJik committed 5 years ago

Verified 8194df8e

add_ctags_to_git_ignore (#2984)

patrickvonplaten committed 5 years ago

Verified 38f5fe9e

Delete untested, broken Model2LSTM (#2968)

sshleifer committed 5 years ago

Verified 129f0604

Correct `special_tokens_mask` when `add_special_tokens=False` (#2965)

LysandreJik committed 5 years ago

Verified 0e84559d

Bart: fix layerdrop and cached decoder_input_ids for generation (#2969)

sshleifer committed 5 years ago

Verified 92487a1d

Add standardized get_vocab method to tokenizers

joeddav committed 5 years ago

Verified c36416e5

fix hardcoded path in examples readme

saippuakauppias committed 5 years ago

cafc4dfc

Update modelcard of bert-base-german-cased

tholor committed 5 years ago

34b4b5a9

Update README.md

mrm8488 committed 5 years ago

7df12d7b

Fix max_length not taken into account when using pad_to_max_length on fast tokenizers (#2961)

mfuntowicz committed 5 years ago

Verified cc6775cd

Older