diff --git a/README.md b/README.md index 64c1cbc469..0b54a10675 100644 --- a/README.md +++ b/README.md @@ -1,26 +1,22 @@ -## Unified and cross-platform CM interface for DevOps, MLOps and MLPerf +# Legacy CM4MLOps repository with DevOps, MLOps and MLPerf automations [![License](https://img.shields.io/badge/License-Apache%202.0-green)](LICENSE.md) [![Powered by CM](https://img.shields.io/badge/Powered_by-MLCommons%20CM-blue)](https://pypi.org/project/cmind). -[![CM script automation features test](https://github.com/mlcommons/cm4mlops/actions/workflows/test-cm-script-features.yml/badge.svg)](https://github.com/mlcommons/cm4mlops/actions/workflows/test-cm-script-features.yml) -[![MLPerf inference bert (deepsparse, tf, onnxruntime, pytorch)](https://github.com/mlcommons/cm4mlops/actions/workflows/test-mlperf-inference-bert-deepsparse-tf-onnxruntime-pytorch.yml/badge.svg)](https://github.com/mlcommons/cm4mlops/actions/workflows/test-mlperf-inference-bert-deepsparse-tf-onnxruntime-pytorch.yml) -[![MLPerf inference MLCommons C++ ResNet50](https://github.com/mlcommons/cm4mlops/actions/workflows/test-mlperf-inference-mlcommons-cpp-resnet50.yml/badge.svg)](https://github.com/mlcommons/cm4mlops/actions/workflows/test-mlperf-inference-mlcommons-cpp-resnet50.yml) -[![MLPerf inference ABTF POC Test](https://github.com/mlcommons/cm4mlops/actions/workflows/test-mlperf-inference-abtf-poc.yml/badge.svg)](https://github.com/mlcommons/cm4mlops/actions/workflows/test-mlperf-inference-abtf-poc.yml) - -# CM4MLOps repository This repository is powered by the [Collective Mind workflow automation framework](https://github.com/mlcommons/ck/tree/master/cm). +The latest sources are available in [this repository](https://github.com/mlcommons/ck/tree/master/cm4mlops). + Two key automations developed using CM are **Script** and **Cache**, which streamline machine learning (ML) workflows, -including managing Docker runs. Both Script and Cache automations are part of the **cmx4mlops** repository. +including managing Docker runs. Both Script and Cache automations are part of the **cm4mlops** repository. The [CM scripts](https://access.cknowledge.org/playground/?action=scripts), also housed in this repository, consist of hundreds of modular Python-wrapped scripts accompanied by `yaml` metadata, enabling the creation of robust and flexible ML workflows. -- **CM Scripts Documentation**: [https://docs.mlcommons.org/cm4mlops/](https://docs.mlcommons.org/cm4mlops/) -- **CM CLI Documentation**: [https://docs.mlcommons.org/ck/specs/cm-cli/](https://docs.mlcommons.org/ck/specs/cm-cli/) +- **CM Scripts Documentation**: [Browse](https://access.cknowledge.org/playground/?action=scripts) +- **CM CLI Documentation**: [https://docs.mlcommons.org/ck/specs/cm-cli/](https://docs.mlcommons.org/ck/specs/cm-cli) ## License @@ -38,31 +34,22 @@ Grigori Fursin, the cTuning foundation and OctoML donated the CK and CM projects ## Author -[Grigori Fursin](https://cKnowledge.org/gfursin) +[Grigori Fursin](https://cKnowledge.org/gfursin). -We sincerely appreciate all [contributors](https://github.com/mlcommons/ck/blob/master/CONTRIBUTORS.md) +We thank all [contributors](https://github.com/mlcommons/ck/blob/master/CONTRIBUTORS.md) for their invaluable feedback and support! ## Concepts Check our [ACM REP'23 keynote](https://doi.org/10.5281/zenodo.8105339) and the [white paper](https://arxiv.org/abs/2406.16791). -## Test image classification and MLPerf R-GAT inference benchmark via CMX PYPI package - -```bash -pip install cmind -pip install cmx4mlops -cmx run script "python app image-classification onnx" --quiet -cmx run script --tags=run,mlperf,inference,generate-run-cmds,_submission,_short --submitter="MLCommons" --adr.inference-src.tags=_branch.dev --pull_changes=yes --pull_inference_changes=yes --submitter="MLCommons" --hw_name=ubuntu-latest_x86 --model=rgat --implementation=python --backend=pytorch --device=cpu --scenario=Offline --test_query_count=500 --adr.compiler.tags=gcc --category=datacenter --quiet --v --target_qps=1 -``` - ## Test image classification and MLPerf R-GAT inference benchmark via CMX GitHub repo ```bash -pip uninstall cmx4mlops pip install cmind -cmx pull repo mlcommons@ck --dir=cmx4mlops/cmx4mlops +cmx pull repo mlcommons@ck --dir=cm4mlops/cm4mlops cmx run script "python app image-classification onnx" --quiet +cmx run script "run-mlperf inference _performance-only _short" --model=resnet50 --precision=float32 --backend=onnxruntime --scenario=Offline --device=cpu --env.CM_SUDO_USER=no --quiet cmx run script --tags=run,mlperf,inference,generate-run-cmds,_submission,_short --submitter="MLCommons" --adr.inference-src.tags=_branch.dev --pull_changes=yes --pull_inference_changes=yes --submitter="MLCommons" --hw_name=ubuntu-latest_x86 --model=rgat --implementation=python --backend=pytorch --device=cpu --scenario=Offline --test_query_count=500 --adr.compiler.tags=gcc --category=datacenter --quiet --v --target_qps=1 ``` diff --git a/automation/script/module.py b/automation/script/module.py index b08875892d..3ec3bc2e0d 100644 --- a/automation/script/module.py +++ b/automation/script/module.py @@ -6558,7 +6558,9 @@ def dump_repro(repro_prefix, rr, run_state): cm_output['version_info'] = version_info if rr['return'] == 0: + # See https://cTuning.org/ae + cm_output['acm_ctuning_repro_badge_available'] = True cm_output['acm_ctuning_repro_badge_functional'] = True diff --git a/script/app-image-classification-torch-py/src/pytorch_classify_preprocessed.py b/script/app-image-classification-torch-py/src/pytorch_classify_preprocessed.py index 863b3a6513..07d690127e 100644 --- a/script/app-image-classification-torch-py/src/pytorch_classify_preprocessed.py +++ b/script/app-image-classification-torch-py/src/pytorch_classify_preprocessed.py @@ -72,7 +72,7 @@ def main(): path_to_model_pth = os.environ['CM_ML_MODEL_FILE_WITH_PATH'] model = models.resnet50(pretrained=False) - model.load_state_dict(torch.load(path_to_model_pth)) + model.load_state_dict(torch.load(path_to_model_pth, weights_only=False)) model.eval()