huggingface/optimum

Pull Requests Commits

Set version number

lewtun committed 3 years ago

249ab717

Add version number to doc build (#168)

lewtun committed 3 years ago

Verified 800e01c4

echarlaix committed 3 years ago

fd6c0137

Improve Documentation (#170)

philschmid committed 3 years ago

Verified 5653a167

Fix rouge keyerror (#167)

JingyaHuang committed 3 years ago

Verified 37a0d86b

[ORTTrainer] Fix CI of ORTTrainer (#165)

JingyaHuang committed 3 years ago

Verified 3a284468

[fix] Warning when loading ORTModels with pipelines of not supported class type (#164)

philschmid committed 3 years ago

Verified 3b522e4d

Ensure attention mask is on same device as inputs IDs during text generation (#162)

lewtun committed 3 years ago

Verified c9761991

[Patch] Add loss for ORT inference (#152)

JingyaHuang committed 3 years ago

Verified 1c4b5b1c

Add a top level init (#160)

fxmarty committed 3 years ago

Verified f9a87e07

Clean workflows (#158)

echarlaix committed 3 years ago

Verified 81b85fa6

Add ORT inference (#113)

philschmid committed 3 years ago

Verified a31e59ed

Fix import (#159)

echarlaix committed 3 years ago

Verified 74172026

Add deepspeed configuration for upcoming tests (#157)

JingyaHuang committed 3 years ago

Verified daeb496d

echarlaix committed 3 years ago

6cb44f08

Fix pretrained ort configuration loading (#156)

echarlaix committed 3 years ago

Verified 73f1ee9d

Update setup (#155)

echarlaix committed 3 years ago

Verified 44794cc2

Add optimum-habana to setup.py and README/doc (#153)

regisss committed 3 years ago

Verified c7efe499

Add the ability to choose the ONNX runtime execution provider in `ORTModel` (#137)

fxmarty committed 3 years ago

Verified 2742bf14

Add ORTOptimizer ELECTRA models support (#150)

echarlaix committed 3 years ago

Verified d36d6ee4

Segment generate dummy inputs to smaller cases (#149)

JingyaHuang committed 3 years ago

Verified 4b1a82a3

Create ONNX configurations with loss as output (#142)

JingyaHuang committed 3 years ago

Verified 8a00085a

Patch release 1.1.0 (#135)

JingyaHuang committed 3 years ago

Verified dfe330a2

ONNX weight sharing (#114)

michaelbenayoun committed 3 years ago

Verified 1e972bf0

Fix readme quickstart indentation (#134)

echarlaix committed 3 years ago

Verified 5dd4c955

Update Dockerfile to be compatible with onnxruntime-training 1.11.0 (#131)

JingyaHuang committed 3 years ago

Verified d470a1de

Add quantization preprocess to include fully connected layer nodes (#133)

echarlaix committed 3 years ago

Verified 913e3be9

Change list element type initializing the auto quantization config (#132)

echarlaix committed 3 years ago

Verified ccfcac16

echarlaix committed 3 years ago

4374af25

[ORTTrainer] Fix to be compatible with onnxruntime 1.11.0 (#124)

JingyaHuang committed 3 years ago

Verified a1dd179d

Older