Update readme, inference submission cleanups (#117)

* Support docker for mixtral

* Added option to automatically submit mlperf inference results while running the submission checker

* Cleanups for get-mlperf-inference-src

* Create CONTRIBUTORS.md

* Update README.md

CONTRIBUTORS.md

@@ -0,0 +1,43 @@
+# Contributors
+
+Thank you for your interest in contributing to **MLPerf Automations**! We welcome contributions that help improve the project and expand its functionality.  
+
+---
+
+## How to Become a Contributor
+
+We value all contributions, whether they are code, documentation, bug reports, or feature suggestions. If you contribute **more than 50 lines of code** (including tests and documentation), you will be officially recognized as a project contributor.  
+
+**Note:** Trivial contributions, such as minor typo fixes or small formatting changes, will not count toward the 50-line threshold.  
+
+To contribute:  
+1. **Fork** the repository.  
+2. **Create** a new branch for your feature or bug fix.  
+3. **Submit** a pull request (PR) describing your changes.
+Please see [here](CONTRIBUTING.md) for further guidelines for official contribution to any MLCommons repository.
+
+---
+
+## Contributor Recognition
+
+Once your contribution exceeds 50 lines of code (in total), we will:  
+- Add your name to this `CONTRIBUTORS.md` file.  
+- Highlight your contribution in the next release notes.  
+- Grant you access to suggest and vote on new features.  
+
+---
+
+## Current Contributors
+
+- **Grigori Fursin** - *Initial Development, CLI workflow support via CMind, Added core automation features*
+- **Arjun Suresh** - *Initial Development, Added core automation features* 
+- **Anandhu Sooraj** - *Added multiple CM scripts for MLPerf Inference*
+- **Thomaz Zhu** - *Added CPP implementation for MLPerf Inference Onnxruntime*
+- **Sahil Avaran** - *Adding logging support in MLPerf script automation* 
+- **[Your Name Here]** - This could be you! 🎉  
+
+---
+
+We believe in collaborative growth, and every contribution makes a difference. Feel free to reach out by opening an issue if you have any questions or ideas.  
+
+Happy Coding! 🚀

README.md

@@ -1,13 +1,65 @@
 # MLPerf Automations and Scripts
 
-This repository contains the automations and scripts used to run MLPerf benchmarks, primarily focusing on MLPerf inference benchmarks. The automations used here are largely based on and extended from the [Collective Mind script automations](https://github.com/mlcommons/cm4mlops/tree/main/automation/script).
+[![License](https://img.shields.io/badge/License-Apache%202.0-green)](LICENSE.md)
+[![Downloads](https://static.pepy.tech/badge/cm4mlops)](https://pepy.tech/project/cm4mlops)
+[![CM Script Automation Test](https://github.com/mlcommons/mlperf-automations/actions/workflows/test-cm-script-features.yml/badge.svg)](https://github.com/mlcommons/mlperf-automations/actions/workflows/test-cm-script-features.yml)
+[![MLPerf Inference ABTF POC Test](https://github.com/mlcommons/mlperf-automations/actions/workflows/test-mlperf-inference-abtf-poc.yml/badge.svg)](https://github.com/mlcommons/mlperf-automations/actions/workflows/test-mlperf-inference-abtf-poc.yml)
 
+Welcome to the **MLPerf Automations and Scripts** repository! This repository provides tools, automations, and scripts to facilitate running MLPerf benchmarks, with a primary focus on **MLPerf Inference benchmarks**.
 
-## Collective Mind (CM) Automations
+The automations build upon and extend the powerful [Collective Mind (CM) script automations](https://github.com/mlcommons/cm4mlops/tree/main/automation/script) to streamline benchmarking and workflow processes.
 
-**CM (Collective Mind)** is a Python package with a CLI and API designed to create and manage automations. Two key automations developed using CM are **Script** and **Cache**, which streamline ML workflows, including managing Docker runs.
+---
 
+## 🚀 Key Features
+- **Automated Benchmarking** – Simplifies running MLPerf Inference benchmarks with minimal manual intervention.
+- **Modular and Extensible** – Easily extend the scripts to support additional benchmarks and configurations.
+- **Seamless Integration** – Compatible with Docker, cloud environments, and local machines.
+- **Collective Mind (CM) Integration** – Utilizes the CM framework to enhance reproducibility and automation.
 
-## License
+---
 
-[Apache 2.0](LICENSE.md)
+## 🧰 Collective Mind (CM) Automations
+
+The **Collective Mind (CM)** framework is a Python-based package offering both CLI and API support for creating and managing automations. CM automations enhance ML workflows by simplifying complex tasks such as Docker container management and caching.
+
+### Core Automations
+- **Script Automation** – Automates script execution across different environments.
+- **Cache Management** – Manages reusable cached results to accelerate workflow processes.
+
+Learn more about CM in the [CM4MLOps documentation](https://github.com/mlcommons/cm4mlops).
+
+---
+
+## 🤝 Contributing
+We welcome contributions from the community! To contribute:
+1. Submit pull requests (PRs) to the **`dev`** branch.
+2. Review our [CONTRIBUTORS.md](here) for guidelines and best practices.
+3. Explore more about MLPerf Inference automation in the official [MLPerf Inference Documentation](https://docs.mlcommons.org/inference/).
+
+Your contributions help drive the project forward!
+
+---
+
+## 📰 News
+Stay tuned for upcoming updates and announcements.
+
+---
+
+## 📄 License
+This project is licensed under the [Apache 2.0 License](LICENSE.md).
+
+---
+
+## 💡 Acknowledgments and Funding
+This project is made possible through the generous support of:
+- [OctoML](https://octoml.ai)
+- [cKnowledge.org](https://cKnowledge.org)
+- [cTuning Foundation](https://cTuning.org)
+- [MLCommons](https://mlcommons.org)
+
+We appreciate their contributions and sponsorship!
+
+---
+
+Thank you for your interest and support in MLPerf Automations and Scripts!

script/generate-mlperf-inference-submission/_cm.yaml

@@ -69,6 +69,7 @@ input_mapping:
   device: CM_MLPERF_DEVICE
   division: CM_MLPERF_SUBMISSION_DIVISION
   duplicate: CM_MLPERF_DUPLICATE_SCENARIO_RESULTS
+  extra_checker_args: CM_MLPERF_SUBMISSION_CHECKER_EXTRA_ARG 
   hw_name: CM_HW_NAME
   hw_notes_extra: CM_MLPERF_SUT_HW_NOTES_EXTRA
   infer_scenario_results: CM_MLPERF_DUPLICATE_SCENARIO_RESULTS

script/get-ml-model-mixtral/_cm.yaml

@@ -6,6 +6,8 @@ category: AI/ML models
 env:
   CM_ML_MODEL_DATASET: ''
   CM_ML_MODEL_WEIGHT_TRANSFORMATIONS: 'no'
+docker:
+  real_run: False
 input_mapping:
   checkpoint: MIXTRAL_CHECKPOINT_PATH
 new_env_keys:

script/get-mlperf-inference-src/_cm.yaml

@@ -50,8 +50,6 @@ prehook_deps:
     _submodules.:
     - CM_GIT_SUBMODULES
 print_env_at_the_end_disabled:
-  CM_MLPERF_INFERENCE_CONF_PATH: Path to the MLPerf inference benchmark configuration
-    file
   CM_MLPERF_INFERENCE_SOURCE: Path to MLPerf inference benchmark sources
 tags:
 - get
@@ -154,31 +152,33 @@ versions:
       CM_MLPERF_LAST_RELEASE: v2.1
       CM_TMP_GIT_CHECKOUT: v2.1
   r3.0:
-    adr:
+    ad:
       inference-git-repo:
         tags: _tag.v3.0
     env:
       CM_MLPERF_LAST_RELEASE: v3.0
       CM_TMP_GIT_CHECKOUT: ''
   r3.1:
-    adr:
+    ad:
       inference-git-repo:
         tags: _tag.v3.1
     env:
       CM_MLPERF_LAST_RELEASE: v3.1
-      CM_TMP_GIT_CHECKOUT: ''
+      CM_GIT_CHECKOUT_TAG: 'v3.1'
   r4.0:
-    adr:
+    ad:
       inference-git-repo:
         tags: _tag.v4.0
     env:
       CM_MLPERF_LAST_RELEASE: v4.0
+      CM_GIT_CHECKOUT_TAG: 'v4.0'
   r4.1:
-    adr:
+    ad:
       inference-git-repo:
         tags: _tag.v4.1
     env:
       CM_MLPERF_LAST_RELEASE: v4.1
+      CM_GIT_CHECKOUT_TAG: 'v4.1'
   r5.0:
     env:
       CM_MLPERF_LAST_RELEASE: v5.0

script/get-mlperf-inference-src/customize.py

@@ -41,7 +41,8 @@ def preprocess(i):
     # if not try to assign the values specified in version parameters,
     # if version parameters does not have the value to a parameter, set the
     # default one
-    if env.get('CM_GIT_CHECKOUT', '') == '':
+    if env.get('CM_GIT_CHECKOUT', '') == '' and env.get(
+            'CM_GIT_CHECKOUT_TAG', '') == '':
         if env.get('CM_TMP_GIT_CHECKOUT', '') != '':
             env["CM_GIT_CHECKOUT"] = env["CM_TMP_GIT_CHECKOUT"]
         else:

script/run-mlperf-inference-submission-checker/_cm.yaml

@@ -36,6 +36,7 @@ deps:
   tags: preprocess,mlperf,inference,submission
 input_mapping:
   extra_args: CM_MLPERF_SUBMISSION_CHECKER_EXTRA_ARGS
+  extra_checker_args: CM_MLPERF_SUBMISSION_CHECKER_EXTRA_ARGS
   extra_model_benchmark_map: CM_MLPERF_EXTRA_MODEL_MAPPING
   input: CM_MLPERF_INFERENCE_SUBMISSION_DIR
   power: CM_MLPERF_POWER
@@ -50,6 +51,7 @@ input_mapping:
   src_version: CM_MLPERF_SUBMISSION_CHECKER_VERSION
   submission_dir: CM_MLPERF_INFERENCE_SUBMISSION_DIR
   submitter: CM_MLPERF_SUBMITTER
+  submitter_id: CM_MLPERF_SUBMITTER_ID
   tar: CM_TAR_SUBMISSION_DIR
 post_deps:
 - enable_if_env:
@@ -66,6 +68,12 @@ post_deps:
     CM_TAR_SUBMISSION_DIR:
     - 'yes'
   tags: run,tar
+- enable_if_env:
+    CM_SUBMITTER_ID:
+    - 'yes'
+  tags: submit,mlperf,results,_inference
+  env:
+    CM_MLPERF_SUBMISSION_FILE: <<<MLPERF_INFERENCE_SUBMISSION_TAR_FILE>>>
 tags:
 - run
 - mlc