NielsRogge/transformers

Pull Requests Commits

support wmt21 tokenizer in m2m100 tokenizer (#14376)

patil-suraj committed 4 years ago

Verified 31659304

Use `AlbertConverter` for FNet instead of using FNet's own converter (#14365)

qqaatw committed 4 years ago

Verified 280a811e

[Wav2Vec2 Example] Improve fine-tuning script (#14373)

patrickvonplaten committed 4 years ago

Verified 55f49c5f

fix docs (#14377)

patil-suraj committed 4 years ago

Verified 21546e59

Adding support for raw python `generator` in addition to `Dataset` for pipelines (#14352)

Narsil committed 4 years ago

Verified ed5d1551

fix --gradient_checkpointing (#13964)

stas00 committed 4 years ago

Verified 77262ef7

fix loading flax bf16 weights in pt (#14369)

patil-suraj committed 4 years ago

Verified 3d607df8

Fixing requirements for TF LM models and use correct model mappings (#14372)

Rocketknight1 committed 4 years ago

Verified 7f20bf0d

Experimenting with adding proper get_config() and from_config() methods (#14361)

Rocketknight1 committed 4 years ago

Verified 4c35c8d8

pass params to encode (#14370)

patil-suraj committed 4 years ago

Verified b1dbdf22

Fix Flax params dtype (#13098)

patil-suraj committed 4 years ago

Verified e92190c0

solve the port conflict (#14362)

stas00 committed 4 years ago

Verified 1c76a516

Fix list index out of range when padding nested empty lists (#13876)

qqaatw committed 4 years ago

Verified 9e37c5cd

enhance rewrite state_dict missing _metadata (#14348)

changwangss committed 4 years ago

Verified bec02ff2

Add notebook INC quantization for text classification tasks (#14293)

echarlaix committed 4 years ago

Verified 2b0d9389

Fix fast tokenization problems (#13930)

qqaatw committed 4 years ago

Verified ea163d09

Adding some quality of life for `pipeline` function. (#14322)

Narsil committed 4 years ago

Verified 5c153079

`BatchFeature`: Convert `List[np.ndarray]` to `np.ndarray` before converting to pytorch tensors (#14306)

eladsegal committed 4 years ago

Verified 321eb562

Support for TF >= 2.7 (#14345)

sgugger committed 4 years ago

Verified 46d0cdae

[Bert2Bert] allow bert2bert + relative embeddings (#14324)

patrickvonplaten committed 4 years ago

Verified e81d8d7f

Rewrite guides for fine-tuning with Datasets (#13923)

stevhliu committed 4 years ago

Verified e4d8f517

bump flax version (#14343)

patil-suraj committed 4 years ago

Verified 85a4bda4

remove test_model_various_embeddings (#14341)

ydshieh committed 4 years ago

Verified babd0b9a

Update Seq2Seq QA example script to use SQuAD metric. (#14335)

karthikrangasai committed 4 years ago

Verified 4f24058c

Add TFViTModel (#13778)

ydshieh committed 4 years ago

Verified be4a6c64

Correct order of overflowing tokens for LayoutLmV2 tokenizer (#13495)

Apoorvgarg-creator committed 4 years ago

Verified 6326aa4b

Add FlaxVisionEncoderDecoderModel (#13359)

ydshieh committed 4 years ago

Verified 95b3ec3b

Small change to Wav2Vec2 model to support Tensor-Parallelism with DeepSpeed (#14298)

RezaYazdaniAminabadi committed 4 years ago

Verified a5030122

[deepspeed] Enable multiple test runs on single box, defer to DS_TEST_PORT if set (#14331)

jeffra committed 4 years ago

Verified d0e96c6d

Expand dynamic supported objects to configs and tokenizers (#14296)

sgugger committed 4 years ago

Verified dfb00bf6

Newer Older