Skip to content

Cleanups to MLPerf inference preprocess script #496

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.

Sign up for GitHub

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

Merged
arjunsuresh merged 5 commits into from
Nov 7, 2024
Merged

Cleanups to MLPerf inference preprocess script #496

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
21425ef
Preprocess mlperf inference submission, code cleanup
arjunsuresh Nov 7, 2024
b40d0c5
Merge branch 'mlcommons:mlperf-inference' into mlperf-inference
arjunsuresh Nov 7, 2024
0c68370
Update test-mlperf-inference-intel
arjunsuresh Nov 7, 2024
ea39555
Update test-mlperf-inference-intel
arjunsuresh Nov 7, 2024
7a8a152
Cleanups for MLPerf inference preprocess script, use inference dev br…
arjunsuresh Nov 7, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions .github/workflows/test-mlperf-inference-intel
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@ name: MLPerf Inference Intel implementations

on:
schedule:
- cron: "49 2 * * *" #to be adjusted
- cron: "54 14 * * *" #to be adjusted

jobs:
build_nvidia:
Expand All @@ -21,5 +21,5 @@ jobs:
source gh_action/bin/activate
export CM_REPOS=$HOME/GH_CM
pip install --upgrade cm4mlops
cm run script --tags=run-mlperf,inference,_all-scenarios,_submission,_full,_r4.1-dev --preprocess_submission=yes --execution_mode=valid --pull_changes=yes --pull_inference_changes=yes --model=${{ matrix.model }} --submitter="MLCommons" --hw_name=RTX4090x2 --implementation=intel --backend=pytorch --category=datacenter --division=open --scenario=Offline --docker_dt=yes --docker_it=no --docker_cm_repo=gateoverflow@cm4mlops --adr.compiler.tags=gcc --device=cpu --results_dir=$HOME/gh_action_results --submission_dir=$HOME/gh_action_submissions --clean --docker --quiet
cm run script --tags=run-mlperf,inference,_all-scenarios,_submission,_full,_r4.1-dev --preprocess_submission=yes --execution_mode=valid --pull_changes=yes --pull_inference_changes=yes --model=${{ matrix.model }} --submitter="MLCommons" --hw_name=IntelSPR.24c --implementation=intel --backend=pytorch --category=datacenter --division=open --scenario=Offline --docker_dt=yes --docker_it=no --docker_cm_repo=gateoverflow@cm4mlops --adr.compiler.tags=gcc --device=cpu --results_dir=$HOME/gh_action_results --submission_dir=$HOME/gh_action_submissions --clean --docker --quiet
cm run script --tags=push,github,mlperf,inference,submission --repo_url=https://github.com/gateoverflow/mlperf_inference_unofficial_submissions_v5.0 --repo_branch=main --commit_message="Results from GH action on SPR.24c" --quiet --submission_dir=$HOME/gh_action_submissions --hw_name=IntelSPR.24c
1 change: 1 addition & 0 deletions script/preprocess-mlperf-inference-submission/_cm.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@
}
],
"input_mapping": {
"input": "CM_MLPERF_INFERENCE_SUBMISSION_DIR",
"submission_dir": "CM_MLPERF_INFERENCE_SUBMISSION_DIR",
"submitter": "CM_MLPERF_SUBMITTER"
},
Expand Down
14 changes: 9 additions & 5 deletions script/preprocess-mlperf-inference-submission/customize.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,17 +11,21 @@ def preprocess(i):
submission_dir = env.get("CM_MLPERF_INFERENCE_SUBMISSION_DIR", "")

if submission_dir == "":
print("Please set CM_MLPERF_INFERENCE_SUBMISSION_DIR")
print("Please set --env.CM_MLPERF_INFERENCE_SUBMISSION_DIR")
return {'return': 1, 'error':'CM_MLPERF_INFERENCE_SUBMISSION_DIR is not specified'}

submitter = env.get("CM_MLPERF_SUBMITTER", "cTuning")
submission_processed = submission_dir + "_processed"
if not os.path.exists(submission_dir):
print("Please set --env.CM_MLPERF_INFERENCE_SUBMISSION_DIR to a valid submission directory")
return {'return': 1, 'error':'CM_MLPERF_INFERENCE_SUBMISSION_DIR is not existing'}

submission_dir = submission_dir.rstrip(os.path.sep)
submitter = env.get("CM_MLPERF_SUBMITTER", "MLCommons")
submission_processed = f"{submission_dir}_processed"

if os.path.exists(submission_processed):
print(f"Cleaning {submission_processed}")
shutil.rmtree(submission_processed)

os.system("rm -rf " + submission_dir + "_processed")

CMD = env['CM_PYTHON_BIN'] + " '" + os.path.join(env['CM_MLPERF_INFERENCE_SOURCE'], "tools", "submission",
"preprocess_submission.py") + "' --input '" + submission_dir + "' --submitter '" + submitter + "' --output '" + submission_processed + "'"
env['CM_RUN_CMD'] = CMD
Expand Down
4 changes: 4 additions & 0 deletions script/run-mlperf-inference-app/_cm.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -264,6 +264,8 @@ variations:
tags: _size.50,_with-sample-ids
nvidia-preprocess-data:
extra_cache_tags: "scc24-base"
inference-src:
tags: _branch.dev
deps:
- tags: clean,nvidia,scratch,_sdxl,_downloaded-data
extra_cache_rm_tags: scc24-main
Expand All @@ -278,6 +280,8 @@ variations:
tags: _size.500,_with-sample-ids
nvidia-preprocess-data:
extra_cache_tags: "scc24-main"
inference-src:
tags: _branch.dev
env:
CM_MLPERF_SUT_NAME_RUN_CONFIG_SUFFIX4: scc24-main
CM_DOCKER_IMAGE_NAME: scc24
Expand Down