-
Notifications
You must be signed in to change notification settings - Fork 712
Arm: Create script for model testing and split run.sh example script #8460
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Arm: Create script for model testing and split run.sh example script #8460
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/8460
Note: Links to docs will display an error until the docs builds have been completed. ❌ 1 New Failure, 1 Pending, 1 Unrelated FailureAs of commit 34738e9 with merge base f0ef51c ( NEW FAILURE - The following job has failed:
BROKEN TRUNK - The following job failed but were present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
@digantdesai there is 2 files outside arm folders are they OK to merge? They are still kind of Arm files, related to arm github testing and the arm delegate guide. 😊 If/when this goes in there is a second patch waiting to be uploaded that add BundleIO testing to this flow so we get input/output verification on our FVP model tests 😊 Edit: Got feedback in other channels by Digant that it was OK to merge this. |
oscarandersson8218
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to me. CI failure should be unrelated.
3589d57 to
f2cb44a
Compare
Create model test script and break up run.sh into smaller scripts that can be used from both places. This makes it possible to run things in run.sh in separate steps like this: 1. Build needed libs backends/arm/scripts/build_executorch.sh backends/arm/scripts/build_portable_kernels.sh --portable_kernels=<OPS> backends/arm/scripts/build_quantized_ops_aot_lib.sh 2. Build <PTE FILE> python3 -m examples.arm.aot_arm_compiler --target=ethos-u85-128 --delegate --quantize --so_library=<PATH>/libquantized_ops_aot_lib.so --model_name=<MODEL> 3. Build target executable <ELF FILE> backends/arm/scripts/build_executorch_runner.sh --pte=<PTE_FILE> --target=ethos-u55-128 4. Test target executable in FVP backends/arm/scripts/run_fvp.sh --elf=<ELF_FILE> --target=ethos-u85-128 Signed-off-by: Zingo Andersen <zingo.andersen@arm.com> Change-Id: I0aefad9e727bc7265e045356d0440d35583957d7
f2cb44a to
34738e9
Compare
|
Fails are unrealted |
Create model test script and break up run.sh into smaller scripts that can be used from both places.
New model test script can be used like this:
$ python3 backends/arm/test/test_model.py --build_libs
$ python3 backends/arm/test/test_model.py --target=ethos-u55-64 --model=lstm
$ python3 backends/arm/test/test_model.py --target=ethos-u85-256 --model=mul
This change also makes it possible to run the things in run.sh in separate steps like this:
$ backends/arm/scripts/build_executorch.sh
$ backends/arm/scripts/build_portable_kernels.sh --portable_kernels=OPS
$ backends/arm/scripts/build_quantized_ops_aot_lib.sh
$ python3 -m examples.arm.aot_arm_compiler --target=ethos-u85-128 --delegate --quantize --so_library=PATH/libquantized_ops_aot_lib.so --model_name=MODEL
$ backends/arm/scripts/build_executorch_runner.sh --pte=PTE_FILE --target=ethos-u85-128
$ backends/arm/scripts/run_fvp.sh --elf=ELF_FILE --target=ethos-u85-128
cc @digantdesai @freddan80 @per @oscarandersson8218