- 28 Oct, 2022 2 commits
- 27 Oct, 2022 17 commits
-
-
charlie authored
-
charlie authored
-
charlie authored
-
charlie authored
Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast
-
charlie authored
Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_ref_multibroadcast
-
charlie authored
-
charlie authored
-
charlie authored
Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast
-
Charlie Lin authored
-
kahmed10 authored
updated GPU pad to now use JIT version. added range functions for JIT kernels.
-
charlie authored
-
charlie authored
Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast
-
charlie authored
-
charlie authored
Merge branch 'dyn_ref_multibroadcast' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_broadcast
-
charlie authored
-
charlie authored
-
charlie authored
-
- 26 Oct, 2022 7 commits
-
-
charlie authored
-
Brian Pickrell authored
Fixes an observed regression error on certain Frozen Protobuf models due to PR 1280
-
kahmed10 authored
use_dynamic_same_auto_pad was removed from convolution, but the driver models still retain the fields. This PR regenerates the files so that they are compatible again.
-
charlie authored
-
charlie authored
-
charlie authored
-
charlie authored
-
- 25 Oct, 2022 2 commits
-
-
Chris Austen authored
-
charlie authored
-
- 24 Oct, 2022 3 commits
-
-
charlie authored
-
charlie authored
-
jungpark-mlir authored
Reiterate the assertion on the standard shape but relax it for the multibroadcast ops deliberately inserted to explicit the broadcast.
-
- 21 Oct, 2022 1 commit
-
-
Umang Yadav authored
-
- 20 Oct, 2022 2 commits
- 19 Oct, 2022 3 commits
-
-
charlie authored
-
Charlie Lin authored
Refactor dynamic compute - add a compute_output_shape object that implicitly converts to a new dyn_output or shape object - dyn_output object can handle computing the static output shape of an operator given the input arguments shapes change an operator's compute function to argument compute(const dyn_output& dyn_out, std::vector<argument> args) to use dyn_output object Dynamic ref unary functions - Included these changes to have an example of the refactored dynamic compute being used - Changes to unary base class to handle dynamic shapes - Changed elu and leaky_relu to use unary base class and pointwise JIT
-
Umang Yadav authored
* use find2.0 for the convolution Co-authored-by:
Vasilii Filippov <DrizztDoUrden@users.noreply.github.com> Co-authored-by:
Chris Austen <causten@users.noreply.github.com>
-
- 18 Oct, 2022 2 commits
-
-
charlie authored
also adding limitations on broadcasting dynamic shapes
-
Paul Fultz II authored
* Enable non-standard shape * Use perfdb for non xdlops * Fix transpose+broadcast strides Co-authored-by:jungpark-mlir <jungwook.park@amd.com>
-
- 17 Oct, 2022 1 commit
-
-
charlie authored
Merge branch 'refactor_dynamic_compute' of github.com:ROCmSoftwarePlatform/AMDMIGraphX into dyn_ref_broadcast
-