NielsRogge/transformers

Pull Requests Commits

Improve conversion script

Niels Rogge committed 3 years ago

d13d412f

Improve implementation

Niels Rogge committed 3 years ago

363947d4

Make conversion script work

Niels Rogge committed 3 years ago

3d833565

Make conversion script work for dpt_large

Niels Rogge committed 3 years ago

f65202ee

Add fusion blocks

Niels Rogge committed 3 years ago

35ce75f4

More improvements

Niels Rogge committed 3 years ago

22f7fb60

Niels Rogge committed 3 years ago

94c7c83c

TF generate refactor - past without encoder outputs (#15944)

gante committed 3 years ago

Verified 70203b59

Update TF multiple choice example (#15868)

gante committed 3 years ago

Verified 62d84760

add hf hub to env version command (#15981)

patrickvonplaten committed 3 years ago

Verified ab2f8d12

Fix TFEncoderDecoderModelTest - Pytorch device (#15979)

ydshieh committed 3 years ago

Verified 72983303

Do a pull in case docs were updated during build (#15922)

sgugger committed 3 years ago

Verified f5a080dd

Speedup training by using numpy instead of jnp for batch shuffling (#15963)

yhavinga committed 3 years ago

Verified 91fb62d0

Returning outputs only when asked for for MaskFormer. (#15936)

Narsil committed 3 years ago

Verified ea07064a

[Tests] Fix ViTMAE integration test (#15949)

NielsRogge committed 3 years ago

Verified b19f3e69

Fix LayoutLMv2 test (#15939)

NielsRogge committed 3 years ago

Verified 9879a1d5

Set scale_embedding to False in some TF tests (#15952)

ydshieh committed 3 years ago

Verified 8b9ae455

Update training scripts docs (#15931)

stevhliu committed 3 years ago

Verified 38cc3506

Better error message when inputs are empty

sgugger committed 3 years ago

c87cfd65

Make is_thing_map in Feature Extractor post_process_panoptic_segmentation defaults to all instances (#15954)

FrancescoSaverioZuppichini committed 3 years ago

Verified e9fa7cd5

Fix Embedding Module Bug in Flax Models (#15920)

sanchit-gandhi committed 3 years ago

Verified 2596f95e

Backprop Test for Freeze FlaxWav2Vec2 Feature Encoder (#15938)

sanchit-gandhi committed 3 years ago

Verified 1a62b25c

Support modern list type hints in HfArgumentParser (#15951)

konstantinjdobler committed 3 years ago

Verified 544fd987

remove re-defination of FlaxWav2Vec2ForCTCModule (#15965)

patil-suraj committed 3 years ago

Verified 60b81dfa

[Bug Fix] Beam search example in docs fails & a fix (integrating `max_length` in `BeamScorer.finalize()`) (#15555)

cwkeam committed 3 years ago

Verified ef9c3ca3

made MaskFormerModelTest faster (#15942)

FrancescoSaverioZuppichini committed 3 years ago

Verified 9932ee4b

Move dependency to call method (#15941)

NielsRogge committed 3 years ago

Verified e8efaecb

Constrained Beam Search [*With* Disjunctive Decoding] (#15761)

cwkeam committed 3 years ago

Verified 5c6f57ee

Tests for MaskFormerFeatureExtractor's post_process*** methods (#15929)

FrancescoSaverioZuppichini committed 3 years ago

Verified 040c11f6

Do not change the output from tuple to list - to match PT's version (#15918)

ydshieh committed 3 years ago

Verified f0aacc14

Newer Older