- 30 Nov, 2023 1 commit
-
-
Lakhinder Walia authored
Ref Unique Operator only
-
- 28 Nov, 2023 1 commit
-
-
music-dino authored
Add min and max reduction modes Implement duplicate index handling Add corresponding reference, onnx, and verify tests.
-
- 23 Nov, 2023 1 commit
-
-
Paul Fultz II authored
-
- 22 Nov, 2023 1 commit
-
-
Mirza Halilčević authored
Introduce dilations attribute to pooling operators reference implementation.
-
- 21 Nov, 2023 2 commits
- 17 Nov, 2023 1 commit
-
-
Umang Yadav authored
Handles all 4 Fp8 dtypes listed here : https://onnx.ai/onnx/technical/float8.html Follows saturation/clipping logic from table there as well : https://onnx.ai/onnx/technical/float8.html#cast Only adding fp8e4m3fnuz in MIGraphX IR for now.
-
- 16 Nov, 2023 1 commit
-
-
Artur Wojcik authored
-
- 15 Nov, 2023 5 commits
-
-
shivadbhavsar authored
Reworked the simplify_qdq pass to support: Per-axis quantization (ie. allow 1D scales and zero points) Allow broadcast and transpose ops between dq and quant_op
-
Paul authored
-
Paul authored
-
Paul authored
-
Paul authored
-
- 13 Nov, 2023 1 commit
-
-
Artur Wojcik authored
-
- 10 Nov, 2023 2 commits
-
-
Artur Wojcik authored
-
Artur Wojcik authored
-
- 09 Nov, 2023 1 commit
-
-
Charlie Lin authored
-
- 08 Nov, 2023 1 commit
-
-
Zakor Gyula authored
The inaccuracy was caused by ONNX round requires nearest integer rounding for halway (0.5) cases. std::round rounds away from zero, thus giving wrong results with halfway cases. Replaced std::round with std::nearbyint which uses the correct rounding by default.
-
- 07 Nov, 2023 2 commits
-
-
Zakor Gyula authored
-
Zakor Gyula authored
-
- 03 Nov, 2023 1 commit
-
-
Brian Pickrell authored
-
- 31 Oct, 2023 1 commit
-
-
Charlie Lin authored
-
- 30 Oct, 2023 1 commit
-
-
Charlie Lin authored
-
- 20 Oct, 2023 1 commit
-
-
Zakor Gyula authored
-
- 19 Oct, 2023 1 commit
-
-
Paul Fultz II authored
-
- 17 Oct, 2023 1 commit
-
-
Charlie Lin authored
-
- 11 Oct, 2023 4 commits
-
-
Ted Themistokleous authored
* Fix scatter operator for nonstandard shapes remove standard() shape check for scatter inputs. * Add nostandard input tests for scatter --------- Co-authored-by:Chris Austen <causten@users.noreply.github.com>
-
Artur Wojcik authored
-
Artur Wojcik authored
-
Artur Wojcik authored
-
- 10 Oct, 2023 1 commit
-
-
Michał Gallus authored
-
- 06 Oct, 2023 3 commits
-
-
Artur Wojcik authored
-
Charlie Lin authored
-
Artur Wojcik authored
-
- 04 Oct, 2023 1 commit
-
-
Artur Wojcik authored
-
- 28 Sep, 2023 2 commits
-
-
Charlie Lin authored
-
Umang Yadav authored
MIGraphX verification by default uses normalized RMS error as the basis for the verification. This change adds some logic to allow migraphx to do "np.allclose" type of elementwise verification using atol and rtol. Commit also includes changes to consistently pass "gold" or "expected" results as the second argument for "verify_range()" calls. Default RMS tolerance inside driver is set to 0.001 which IMO is high for FP32 compared to what we had earlier. Need better defaults
-
- 27 Sep, 2023 1 commit
-
-
Ted Themistokleous authored
Modify reshapes to use reshape_lazy for aliasing and then reshape for a reshape copy operation to eliminate contiguous
-
- 24 Sep, 2023 1 commit
-
-
Charlie Lin authored
New compiler pass that simplifies dynamic shapes related operators to their static versions if possible Will normally be used after a split_single_dyn_dim pass
-
- 21 Sep, 2023 1 commit
-
-
Chris Austen authored
* Rectify flipped coordinate_transformation_mode logic in ROIAlign * Handle both opset 10 and 16 versions * Fix version check and clang tidy warning Co-authored-by:Dino Musić <dino.music@htecgroup.com>
-