microsoft/onnxruntime

Pull Requests Commits

Add INT16 and UINT16 compatibility for relu_quantizelinear (#20187)

danyue333 committed 2 years ago

Verified 07b5377f

Support more ops for recompute (#20234)

pengwa committed 2 years ago

Verified 41acd8c5

Fix Mixtral Parity test to keep it consistent with Transformers. (#20210)

AdamLouly committed 2 years ago

Verified 22a61a3c

fix "4bit quantization scales and zeropoint tensor shape" (#19986)

wejoncy committed 2 years ago

Verified 908a76d6

[js/webgpu] Implement com.microsoft.RotaryEmbedding (#20209)

hujiajie committed 2 years ago

Verified 23d3afd4

Improve KE for commandline and programmatically tuning dispatch (#18778)

cloudhan committed 2 years ago

Verified e19c7789

Support seq_len > 64K in rotary embedding cuda kernel (#20204)

gh-yewang committed 2 years ago

Verified cc3faba6

Remove non-existing function call (#19416)

spadacesco committed 2 years ago

Verified 6af02ae0

Update QNN python packages to use QNN SDK version 2.19.2 (#20213)

adrianlizarraga committed 2 years ago

Verified 05d97e8d

Extend time out in Windows GPU packaging jobs (#20207)

mszhanyi committed 2 years ago

Verified 23a5d0a3

Fix HalideIR title in third party notices reference (#20190)

ei-grad committed 2 years ago

Verified a6611409

Bump Sixlabors.ImageSharp from 2.1.1 to 2.1.7 in /csharp/sample/Microsoft.ML.OnnxRuntime.ResNet50v2Sample (#19805)

dependabot[bot] committed 2 years ago

Verified 2a323eb6

Improve the script to add Q, DQ nodes around EPContext node (#20107)

HectorSVC committed 2 years ago

Verified 1ccb164c

fix ConvTranspose 1D (#20194)

guschmue committed 2 years ago

Verified c529e05e

Bump Sixlabors.ImageSharp from 2.1.1 to 2.1.7 in /csharp/sample/Microsoft.ML.OnnxRuntime.FasterRcnnSample (#19806)

dependabot[bot] committed 2 years ago

Verified 4f2d4542

Add onnxruntime/test/run_benchmark.py helper script. (#19234)

edgchen1 committed 2 years ago

Verified 2b307111

[js/rn] Support load external data (#20090)

hans00 committed 2 years ago

Verified 6abfb6b9

NNAPI: Improve MatMul diagnostic output (#19721)

skottmckay committed 2 years ago

Verified f61cca1b

OneDNN/dnnl: Fix filepath after dnnl move (#20086)

tboby committed 2 years ago

Verified 254bdbb1

[Fix] Upload training CUDA daily wheel (#20183)

mszhanyi committed 2 years ago

Verified 4ea54b82

Fix build errors from date/date.h C++20 compatibility (#20139)

afantino951 committed 2 years ago

Verified 7303a90f

Support building Windows CUDA with Ninja (#20176)

mszhanyi committed 2 years ago

Verified dae77e60

[js/webgpu] add validation to workgroup size (#20110)

fs-eire committed 2 years ago

Verified fa1917b8

Export of Openai Whisper with batched prompts (#19854)

shubhambhokare1 committed 2 years ago

Verified be831e1b

#19921 [Dup] LLC Core count calculations updated (#20171)

YUNQIUGUO committed 2 years ago

Verified 19793de1

Add new SessionOptions config entry to disable specific transformers and rules (#20135)

yuslepukhin committed 2 years ago

Verified 12e25380

[TensorRT EP] Address compiler warnings on Windows (#20134)

chilo-ms committed 2 years ago

Verified e9169293

[js/webgpu] Use global id in attention and instance-norm (#20008)

axinging committed 2 years ago

Verified a2998e5d

[java][DML EP] Modifying dml_provider_factory.h so it can compile as a C header file (#20157)

Craigacp committed 2 years ago

Verified 262b6bd3

Update api backward compatibility (#20136)

xiaoyu-work committed 2 years ago

Verified 3979f53a

Newer Older