Modify MIGraphX EP for Accuracy tests#13455
Conversation
|
This pull request fixes 1 alert when merging f0ea9d1 into 70b73af - view on LGTM.com fixed alerts:
|
f0ea9d1 to
224b39b
Compare
b546d4c to
7ca3601
Compare
|
This pull request fixes 1 alert when merging 7ca3601 into ab9ac2a - view on LGTM.com fixed alerts:
|
|
@tianleiwu is there anything else we need for this PR to go through? |
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
| ) | ||
|
|
||
| result["provider"] = "CUDAExecutionProvider" if args.use_gpu else "CPUExecutionProvider" | ||
| result["provider"] = args.provider if use_gpu else "CPUExecutionProvider" |
There was a problem hiding this comment.
Why it cannot be result["provider"] = args.provider
ccee851 to
ebf28e8
Compare
| assert ( | ||
| "CUDAExecutionProvider" in onnxruntime.get_available_providers() | ||
| ), "Please install onnxruntime-gpu package to test GPU inference." | ||
| assert any( |
There was a problem hiding this comment.
You forgot to remove the any() here. isdisjoint have already returned a bool here.
There was a problem hiding this comment.
Ah right. Let me fix this
ebf28e8 to
5789165
Compare
5789165 to
33d5492
Compare
180e5ca to
8301ffc
Compare
8301ffc to
e2aaf77
Compare
e2aaf77 to
7f03036
Compare
| gpu_ep.append("MIGraphXExecutionProvider") | ||
| gpu_ep.append("ROCMExecutionProvider") | ||
|
|
||
| session = onnxruntime.InferenceSession(onnx_model_path, sess_options, providers=gpu_ep, **kwargs) |
Check notice
Code scanning / CodeQL
Unused local variable
|
This pull request introduces 1 alert when merging 7f03036 into 4ca62b9 - view on LGTM.com new alerts:
Heads-up: LGTM.com's PR analysis will be disabled on the 5th of December, and LGTM.com will be shut down ⏻ completely on the 16th of December 2022. It looks like GitHub code scanning with CodeQL is already set up for this repo, so no further action is needed 🚀. For more information, please check out our post on the GitHub blog. |
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
This comment was marked as resolved.
### Description fix a wrong assert condition in benchmark_helper.py (introduced in #13455)
Allows MIGraphX EP to run the following additional tests. Also adds support to get MIGraphX to run eval_squad.py Reference to the Rocm EP changes: microsoft#13306 Co-authored-by: Joseph Groenenboom <joseph.groenenboom@amd.com> Co-authored-by: Ted Themistokleous <tthemist@amd.com>
### Description fix a wrong assert condition in benchmark_helper.py (introduced in #13455)
Allows MIGraphX EP to run the following additional tests. Also adds support to get MIGraphX to run eval_squad.py Reference to the Rocm EP changes: microsoft#13306 Co-authored-by: Joseph Groenenboom <joseph.groenenboom@amd.com> Co-authored-by: Ted Themistokleous <tthemist@amd.com>
### Description fix a wrong assert condition in benchmark_helper.py (introduced in microsoft#13455)
Description
Changes to allow MIGraphX EP while leveraging the work ROCm has done for the EP for getting additional accuracy tests running.
Motivation and Context
Allows MIGraphX EP to run the following additional tests
test_parity_gelu.py
test_parity_layernorm.py
test_parity_huggingface_gpt_attention.py
convert_to_onnx.py
Reference to the Rocm EP changes: #13306
Also adds support to get MIGraphX to run eval_squad.py