microsoft/onnxruntime

Pull Requests Commits

[js/web] make ort-web export compatible with webpack

fs-eire committed 1 year ago

5371650c

Fix an issue that QNN models shared from other session use the session logger from that session (#22170)

HectorSVC committed 1 year ago

Verified b636b275

Add benchmark script for segment anything v2 (#22169)

tianleiwu committed 1 year ago

Verified 171b901e

Add fusion script for segment anything v2 (#22167)

tianleiwu committed 1 year ago

Verified 1431215d

Address ZeroK case for Gemm for CPU and CUDA (#22111)

yuslepukhin committed 1 year ago

Verified fe8a10ca

set CMAKE_SYSTEM_PROCESSOR in xnnpack.cmake (#22155)

mszhanyi committed 1 year ago

Verified 8d2d4078

Fix std::chrono/date conflict for mac builds with C++20 (#22138)

skottmckay committed 1 year ago

Verified d4692835

Fix CUDA reduction ops handling of optional axes input (#22149)

skottmckay committed 1 year ago

Verified da3bd45c

Fix memory access violations in the CPU float16 min and max operators (#22135)

adamreeve committed 1 year ago

Verified f3cbe760

[CPU EP] Refactor MatMulNBits to decouple type implementation (#22140)

fajin-corp committed 1 year ago

Verified b0ef1f39

[qnn ep] fix naming convention of ort-nightly-qnn package (#22157)

jywu-msft committed 1 year ago

Verified c270fe6d

Fix QNN random crash for UT with multi-thread run (#22160)

HectorSVC committed 1 year ago

Verified 03ce996b

Set Transpose Attribute instead for manipulating MatMul Strides (#21927)

AnaghaRaoAMD committed 1 year ago

Verified 73b5c335

Update nuget.exe used in WindowsAI nuget packaging so `readme` property is supported. (#22141)

skottmckay committed 1 year ago

Verified bd60add8

Enable Android 16 KB page size support (#22076)

skottmckay committed 1 year ago

Verified 99ee6eec

[WebNN EP] Use both MLOperandDescriptor.dimensions and MLOperandDescriptor.shape (#22121)

Honry committed 1 year ago

Verified e33b08ea

[QNN EP] set up py packaging pipeline for Linux x64 (#22132)

jywu-msft committed 1 year ago

Verified 944d8738

Implementation of AVX-VNNI-INT8 dot product instructions into MLAS GEMM (#21984)

mguynn-intc committed 1 year ago

Verified d5f6343a

use mac 12 for esrp code sign (#22134)

mszhanyi committed 1 year ago

Verified 560778fd

Add onnx export script for segment anything v2 (#22119)

tianleiwu committed 1 year ago

Verified a9740d6f

[DML EP] Add QDQ+MatMul fusion into MatMulNBits (#22114)

PatriceVignola committed 1 year ago

Verified 05acfb90

[QNN EP] Build Python 3.12 wheel for Windows ARM64 (#22118)

adrianlizarraga committed 1 year ago

Verified b8dae685

Update q4common.h to include the missing header (#21786)

csukuangfj committed 1 year ago

Verified c6dc787a

Bump body-parser from 1.20.1 to 1.20.3 in /onnxruntime/test/wasm (#22106)

dependabot[bot] committed 1 year ago

Verified 7e989268

Speedup `CumSum` for large arrays (#22048)

neNasko1 committed 1 year ago

Verified 275eb404

Upgrade XNNPACK to latest version (#22012)

mszhanyi committed 1 year ago

Verified b94ba09e

Update pool to MacOS-13 (#17361)

jchen351 committed 1 year ago

Verified fa68ae2d

[TensorRT EP] Add supportsModelV2 (#22081)

chilo-ms committed 1 year ago

Verified 6dcdc70a

[WebNN EP] Support QuantizeLinear and DequantizeLinear ops (#22097)

Honry committed 1 year ago

Verified 9786909a

[js/webgpu] Replace array with string in transpose perm (#21930)

axinging committed 1 year ago

Verified afd642a1

Older