huggingface/optimum

Pull Requests Commits

Release: v1.23.0

echarlaix committed 1 year ago

16311c59

Remove upper transformers version limit (#2048)

echarlaix committed 1 year ago

Verified 851f04b1

Remove the need for the config to be in the subfolder (#2044)

echarlaix committed 1 year ago

Verified 1b5a63da

Enable ONNX export for transformers 4.45 (#2045)

echarlaix committed 1 year ago

Verified 2c0476ed

Update/Fix Pipeline Mixins and ORT Pipelines (#2021)

IlyasMoutawwakil committed 1 year ago

Verified d3c56cd5

Remove numpy version constraint in setup.py (#2039)

h3110Fr13nd committed 1 year ago

Verified d9754abd

Add Transformers v4.45 support (#2023)

echarlaix committed 1 year ago

Verified 049b00f6

Add ipex to documentation (#2027)

rbrugaro committed 1 year ago

Verified c6b46786

CI - update runner type (#2033)

glegendre01 committed 1 year ago

Verified f7c3a7fa

Added image-to-image task for ORT Pipeline (#2031)

h3110Fr13nd committed 1 year ago

Verified fd638d20

Fix `is_torch_tpu_available` in ORT Trainer (#2028)

regisss committed 1 year ago

Verified 2fb5ea5c

Add Parallel Cross Entropy (#2017)

zhenglongjiepheonix committed 1 year ago

Verified bf1befdf

Disable the exllama on all non-cuda devices. (#2003)

yuanwu2017 committed 1 year ago

Verified 2179d33a

Adding `ORTPipelineForxxx` entrypoints (#1960)

IlyasMoutawwakil committed 1 year ago

Verified ca36fc4f

Fixes detection of CuPy installed with pre-built wheels (#1965)

tcsavage committed 1 year ago

Verified f1b708c4

Dev version (#2022)

echarlaix committed 1 year ago

Verified 26949f58

Add quanto install and instructions (#1976)

dacorvo committed 1 year ago

Verified e604af32

update transformers imports for `deepspeed` and `is_torch_xla_available` (#2012)

Rohan138 committed 1 year ago

Verified 2335ec25

Apply deprecated `evaluation_strategy` (#1819)

muellerzr committed 1 year ago

Verified 29f23f1f

Fix typo in BetterTransformer's overview docs (#2015)

ftnext committed 1 year ago

Verified c0d91117

fix attribute name from `inputs_names` to `input_names` (#2010)

J4BEZ committed 1 year ago

Verified 1de4e252

Fix TFLite tests (#2007)

IlyasMoutawwakil committed 1 year ago

Verified 8cb6832a

Modify token classification processor default dataset args (#2005)

echarlaix committed 1 year ago

Verified bb46ebea

Modify Parallelization Strategy to Make it More General (#1988)

zhenglongjiepheonix committed 1 year ago

Verified ad98dc94

Transformers 4.44 support (#1996)

IlyasMoutawwakil committed 1 year ago

Verified 7cc57e40

Follow up the diffusers task refactoring (#1999)

JingyaHuang committed 1 year ago

Verified 3b558756

Add Param Cache For Recompilation (#2000)

zhenglongjiepheonix committed 1 year ago

Verified 23f8574e

Update Habana extra in setup.py (#1991)

regisss committed 1 year ago

Verified 70bd6725

Fix typos in quantization.mdx (#1989)

aldakata committed 1 year ago

Verified fa2a2986

Automatic Model Parallelism Through FX (#1933)

zhenglongjiepheonix committed 1 year ago

Verified 5eaf91ba

Older