Commits · d2532d0eee52c8e12d05990ce6a7c8bec6480df7 · gaoqiong / MIGraphX

13 Dec, 2023 5 commits
- Format · d2532d0e
  Paul authored Dec 13, 2023
  
  d2532d0e
- Add explicit call to optimize_onnx · 2338326b
  Paul authored Dec 13, 2023
  
  2338326b
- Add cmake for onnx tests · acbedb7d
  Paul authored Dec 13, 2023
  
  acbedb7d
- Lazy reshape fixes (#2505) · a60bdb67
  Paul Fultz II authored Dec 13, 2023
  
  a60bdb67
- Make reshape copy always have standard shape as output (#2508) · b6976b94
  Paul Fultz II authored Dec 13, 2023
  
  b6976b94
12 Dec, 2023 7 commits
- Add DynamicQuantizeLinear op (#2489) · 5fe1b075
  Zakor Gyula authored Dec 12, 2023
  
  5fe1b075
- refactoring quantization passes (#2544) · 7e61114a
  Umang Yadav authored Dec 12, 2023
  
  7e61114a
- Update onnxruntime main 44b58437402b207c8216f3be8c75accb7409be1c (#2538) · b742b528
  github-actions[bot] authored Dec 12, 2023
  
  b742b528
- Add Qlinearconcat op (#2476) · 7e03e055
  Zakor Gyula authored Dec 12, 2023
  
  7e03e055
- Bump rocMLIR 6.1 commit pointer (#2543) · 6e484d77
  Manupa Karunaratne authored Dec 12, 2023
  
  6e484d77
- Add `--fp8` option to quantize models in FP8 inside `migraphx-driver` (#2535) · db3c07fb
  Umang Yadav authored Dec 12, 2023
  
  db3c07fb
- FP8 QuantDot operation (#2506) · aac4e950
  Umang Yadav authored Dec 12, 2023
  
  aac4e950
09 Dec, 2023 1 commit
- Move rocMLIR to tip of release branch (#2539) · 9d2003a2
  Krzysztof Drewniak authored Dec 08, 2023
  
  9d2003a2
07 Dec, 2023 6 commits
- Add Fp8 warnings in finalize (#2532) · 7e53592e
  Umang Yadav authored Dec 07, 2023
```
Enable simplify qdq to work with FP8 types, add warning for fp8 when loading mxr
```
  7e53592e
- Bump rocm-docs-core from 0.30.0 to 0.30.1 in /docs/.sphinx (#2529) · 4c059fa3
  dependabot[bot] authored Dec 07, 2023
  
  4c059fa3
- print ref results (#2530) · fe61d940
  Umang Yadav authored Dec 07, 2023
  
  fe61d940
- Enable simplify qdq to work with FP8 types (#2528) · 22f60c24
  Umang Yadav authored Dec 07, 2023
  
  22f60c24
- ad fp8 warnings (#2531) · dfc18d6c
  Umang Yadav authored Dec 07, 2023
  
  dfc18d6c
- FP8 2D forward convolution using rocMLIR (#2507) · 6a72e8fc
  Umang Yadav authored Dec 06, 2023
  
  6a72e8fc
06 Dec, 2023 6 commits
- Enable standalone convolution for mlir by default (#2503) · a09dc502
  Paul Fultz II authored Dec 05, 2023
  
  a09dc502
- Enable running individual passes in the driver (#2413) · 93f8c81d
  Paul Fultz II authored Dec 05, 2023
  
  93f8c81d
- Improve dpp reductions on navi (#2439) · 8c73c72e
  Paul Fultz II authored Dec 05, 2023
  
  8c73c72e
- Set MIGRAPHX_USE_COMPOSABLEKERNEL=1 in cmake (#2518) · 7d4b1719
  turneram authored Dec 05, 2023
  
  7d4b1719
- Device kernels using FP8 (#2510) · 98ef0abb
  Umang Yadav authored Dec 05, 2023
  
  98ef0abb
- Add FP8 rocblas gemm support (#2473) · 6d0b6bcf
  Umang Yadav authored Dec 05, 2023
  
  6d0b6bcf
04 Dec, 2023 3 commits
- [MLIR] Updates needed for general stride support (#2428) · e3e00547
  Krzysztof Drewniak authored Dec 04, 2023
  
  e3e00547
- Remove unused device kernels (#2509) · 609ad966
  Umang Yadav authored Dec 04, 2023
  
  609ad966
- Update onnx proto and onnx parser to handle fp8 types (#2493) · 9c46821c
  Umang Yadav authored Dec 04, 2023
```
Updated onnx.proto files to be same as onnx==1.14.1 that is used by MIGraphX.   Added more switch cases in onnx parser to handle FP8 dtypes.
```
  9c46821c
03 Dec, 2023 1 commit
- Fixes doc generation and cross referencing (#2502) · 02f7405a
  Paul Fultz II authored Dec 03, 2023
  
  02f7405a
02 Dec, 2023 2 commits
- Update onnxruntime main d69842226b47e5336568103541b071447caeb9bf (#2500) · 32f0b028
  github-actions[bot] authored Dec 01, 2023
  
  32f0b028
- Only format certain extensions (#2504) · c6613075
  Paul Fultz II authored Dec 01, 2023
  
  c6613075
01 Dec, 2023 3 commits
- FP8 GPU implementation (#2455) · eafd55de
  Umang Yadav authored Dec 01, 2023
  
  eafd55de
- Index the env vars (#2501) · 785ff7d7
  Paul Fultz II authored Dec 01, 2023
```
Co-authored-by: Chris Austen <causten@users.noreply.github.com>
```
  785ff7d7
- Add a tuning limit for mlir (#2442) · b47dd32b
  Paul Fultz II authored Dec 01, 2023
  
  b47dd32b
30 Nov, 2023 4 commits
- Bump rocm-docs-core from 0.29.0 to 0.30.0 in /docs/.sphinx (#2492) · fd7e92af
  dependabot[bot] authored Nov 30, 2023
  
  fd7e92af
- Have the accuracy checker print the whole final buffer (#2496) · 799f1c49
  Charlie Lin authored Nov 30, 2023
  
  799f1c49
- [6.1] Add support for dot-(mul)-softmax-dot offloads to MLIR (#2345) · ff485c7a
  Manupa Karunaratne authored Nov 30, 2023
  
  ff485c7a
- Unique Operator. (#2072) · 6d84f7c6
  Lakhinder Walia authored Nov 30, 2023
```
Ref Unique Operator only
```
  6d84f7c6
29 Nov, 2023 2 commits
- Fix agent model and multinomial parsing (#2481) · 44463b94
  Umang Yadav authored Nov 29, 2023
  
  44463b94
- Add find_dependency for TBB (#2483) · 7eee28ce
  Paul Fultz II authored Nov 29, 2023
  
  7eee28ce