Commits · 2d6acab36dfc72346566e4eb0a15b8045c5ed48f · gaoqiong / MIGraphX

"README.md" did not exist on "7339f0b0e2449d7f0fecc88b930d5842f4df26ac"

02 Nov, 2023 6 commits
- python fix · 2d6acab3
  Umang Yadav authored Nov 01, 2023
  
  2d6acab3
- test_gpu_jit working after removing implicit_conversion_op · 4935b575
  Umang Yadav authored Nov 01, 2023
  
  4935b575
- Compiles all right now · 16b5e050
  Umang Yadav authored Oct 31, 2023
  
  16b5e050
- Add type in API, add MIGRAPHX_HIP_HOST_DEVICE · c2b41d9d
  Umang Yadav authored Oct 31, 2023
  
  c2b41d9d
- Format · c75fb295
  Paul authored Nov 02, 2023
  
  c75fb295
- Fix merge conflicts: · e69b4a33
  Paul authored Nov 02, 2023
  
  e69b4a33
31 Oct, 2023 2 commits
- Run simplify_qdq first before optimize_module (#2387) · 254c2a78
  Paul Fultz II authored Oct 31, 2023
  
  254c2a78
- Fix allocate 0 arg version (#2388) · 74fb919a
  Charlie Lin authored Oct 31, 2023
  
  74fb919a
30 Oct, 2023 4 commits
- Remove int8x4 format completely (#2373) · 22bb777f
  Umang Yadav authored Oct 30, 2023
  
  22bb777f
- fix tests and error cases; allow std::optional for streamutils (#2362) · ec83e545
  Charlie Lin authored Oct 29, 2023
  
  ec83e545
- make CK JIT optional (#2383) · 728bea34
  Artur Wojcik authored Oct 30, 2023
  
  728bea34
- use find_package() for SQLite3 (#2384) · 347f805e
  Artur Wojcik authored Oct 30, 2023
  
  347f805e
27 Oct, 2023 1 commit
- Add missing space in perf report (#2368) · 830cde6c
  Paul Fultz II authored Oct 27, 2023
  
  830cde6c
25 Oct, 2023 1 commit
- Fix wrong size check when axes not present for slice (#2270) · 52203fa4
  Attila Dusnoki authored Oct 25, 2023
  
  52203fa4
24 Oct, 2023 1 commit
- Ensure unique module name for MLIR standalone ops (#2360) · d1abf06f
  Paul Fultz II authored Oct 24, 2023
  
  d1abf06f
23 Oct, 2023 1 commit
- Update unit as inference/sec (#2352) · 7604ecf5
  pramenku authored Oct 23, 2023
  
  7604ecf5
21 Oct, 2023 1 commit

Add benchmark tracing (#2354) · 8f9ccb9a

Paul Fultz II authored Oct 20, 2023

 Add tracing to benchmark to show which kernels are running and the time of every kernel

8f9ccb9a

20 Oct, 2023 4 commits
- Add support for select_last_index attribute for ArgMax & ArgMin (#2235) · 6ae4227a
  Zakor Gyula authored Oct 20, 2023
  
  6ae4227a
- CK GEMM Int8 Bug Fixes (#2229) · f47e0b5b
  turneram authored Oct 19, 2023
```
Adds workarounds to avoid passing capture ops and scalar literals from quantization as arguments to ck_gemm.
```
  f47e0b5b
- Fix trilu operator computation logic (#2212) · b8b4630b
  nives-vukovic authored Oct 20, 2023
  
  b8b4630b
- Add MeanVarianceNormalization ONNX parsing (#2255) · 6072b2c4
  music-dino authored Oct 20, 2023
  
  6072b2c4
19 Oct, 2023 2 commits
- Make argument constructor explicit (#2346) · 07848b28
  Paul Fultz II authored Oct 19, 2023
  
  07848b28
- Add flag to accept non-uniform WG sizes (#2167) · 581b1b5f
  Umang Yadav authored Oct 19, 2023
```
* Disable -Wunsafe-buffer-usage when compiling gpu code
```
  581b1b5f
18 Oct, 2023 8 commits
- Add support for Shrink ONNX operator (#2240) · c3990622
  Zakor Gyula authored Oct 18, 2023
  
  c3990622
- Update header · ab096227
  Paul authored Oct 18, 2023
  
  ab096227
- Update macro · 01490e32
  Paul authored Oct 18, 2023
  
  01490e32
- Format · a61a8ffc
  Paul authored Oct 18, 2023
  
  a61a8ffc
- Fix up timing of kernel · 5a2cfafd
  Paul authored Oct 18, 2023
  
  5a2cfafd
- Fix implicit conversion that is causing perf degration · 94bba2c1
  Paul authored Oct 18, 2023
  
  94bba2c1
- Some more merge fixes · 1fd73bac
  Paul authored Oct 18, 2023
  
  1fd73bac
- Change driver verify to check for fp16 and --fp16 (#2334) · 5139b930
  Charlie Lin authored Oct 17, 2023
  
  5139b930
17 Oct, 2023 9 commits
- Format · a42edd9d
  Paul authored Oct 17, 2023
  
  a42edd9d
- Remvoe stray if · 7892b274
  Paul authored Oct 17, 2023
  
  7892b274
- Format · 9ca6c1d8
  Paul authored Oct 17, 2023
  
  9ca6c1d8
- Initialize variables · 070f8c95
  Paul authored Oct 17, 2023
  
  070f8c95
- Add axes (optional) input to Pad (#2178) · 94bda243
  Attila Dusnoki authored Oct 17, 2023
  
  94bda243
- Format · e1ef6e2c
  Paul authored Oct 17, 2023
  
  e1ef6e2c
- Format · e846ae72
  Paul authored Oct 17, 2023
  
  e846ae72
- Fix compilation error · dc71e23d
  Paul authored Oct 17, 2023
  
  dc71e23d
- Add GroupNorm and LayerNorm onnx parsing (#2242) · 52c74f0e
  Attila Dusnoki authored Oct 17, 2023
  
  52c74f0e