Skip to content

generate-mlperf-inference-submission

Automatically generated README for this automation recipe: generate-mlperf-inference-submission

Category: MLPerf benchmark support

License: Apache 2.0

  • Notes from the authors, contributors and users: README-extra

  • CM meta description for this script: _cm.json

  • Output cached? False

Reuse this script in your project

Install MLCommons CM automation meta-framework

Pull CM repository with this automation recipe (CM script)

cm pull repo mlcommons@cm4mlops

cmr "generate submission mlperf mlperf-inference inference mlcommons inference-submission mlperf-inference-submission mlcommons-inference-submission" --help

Run this script

Run this script via CLI
cm run script --tags=generate,submission,mlperf,mlperf-inference,inference,mlcommons,inference-submission,mlperf-inference-submission,mlcommons-inference-submission [--input_flags]
Run this script via CLI (alternative)
cmr "generate submission mlperf mlperf-inference inference mlcommons inference-submission mlperf-inference-submission mlcommons-inference-submission " [--input_flags]
Run this script from Python
import cmind

r = cmind.access({'action':'run'
              'automation':'script',
              'tags':'generate,submission,mlperf,mlperf-inference,inference,mlcommons,inference-submission,mlperf-inference-submission,mlcommons-inference-submission'
              'out':'con',
              ...
              (other input keys for this script)
              ...
             })

if r['return']>0:
    print (r['error'])
Run this script via Docker (beta)
cm docker script "generate submission mlperf mlperf-inference inference mlcommons inference-submission mlperf-inference-submission mlcommons-inference-submission" [--input_flags]

Script flags mapped to environment

  • --analyzer_settings_file=value → CM_MLPERF_POWER_ANALYZER_SETTINGS_FILE_PATH=value
  • --category=value → CM_MLPERF_SUBMISSION_CATEGORY=value
  • --clean=value → CM_MLPERF_CLEAN_SUBMISSION_DIR=value
  • --dashboard=value → CM_MLPERF_DASHBOARD=value
  • --dashboard_wb_project=value → CM_MLPERF_DASHBOARD_WANDB_PROJECT=value
  • --device=value → CM_MLPERF_DEVICE=value
  • --division=value → CM_MLPERF_SUBMISSION_DIVISION=value
  • --duplicate=value → CM_MLPERF_DUPLICATE_SCENARIO_RESULTS=value
  • --hw_name=value → CM_HW_NAME=value
  • --hw_notes_extra=value → CM_MLPERF_SUT_HW_NOTES_EXTRA=value
  • --infer_scenario_results=value → CM_MLPERF_DUPLICATE_SCENARIO_RESULTS=value
  • --power_settings_file=value → CM_MLPERF_POWER_SETTINGS_FILE_PATH=value
  • --preprocess=value → CM_RUN_MLPERF_SUBMISSION_PREPROCESSOR=value
  • --preprocess_submission=value → CM_RUN_MLPERF_SUBMISSION_PREPROCESSOR=value
  • --results_dir=value → CM_MLPERF_INFERENCE_RESULTS_DIR_=value
  • --run_checker=value → CM_RUN_SUBMISSION_CHECKER=value
  • --run_style=value → CM_MLPERF_RUN_STYLE=value
  • --skip_truncation=value → CM_SKIP_TRUNCATE_ACCURACY=value
  • --submission_dir=value → CM_MLPERF_INFERENCE_SUBMISSION_DIR=value
  • --submitter=value → CM_MLPERF_SUBMITTER=value
  • --sw_notes_extra=value → CM_MLPERF_SUT_SW_NOTES_EXTRA=value
  • --tar=value → CM_TAR_SUBMISSION_DIR=value

Default environment

These keys can be updated via --env.KEY=VALUE or env dictionary in @input.json or using script flags.

  • CM_RUN_MLPERF_ACCURACY: on
  • CM_MLPERF_RUN_STYLE: valid

Script output

cmr "generate submission mlperf mlperf-inference inference mlcommons inference-submission mlperf-inference-submission mlcommons-inference-submission " [--input_flags] -j