# cmf | Hewlett Packard Enterprise | Knowledge Base | Instagit

CMF library helps to collect and store information associated with ML pipelines.  It tracks the lineages for artifacts and executions of distributed AI pipelines.  It provides API's to record and query the metadata associated with ML pipelines. The framework adopts a data first approach and all artifacts recorded in the framework are versioned and identified by the content hash.

GitHub Stars: 34

Repository: https://github.com/hewlettpackard/cmf

---

## Articles

### [How to Implement Custom Storage Backends for CMF Artifacts: A Complete Guide](/hewlettpackard/cmf/how-to-implement-custom-storage-backends-for-cmf-artifacts)

Learn how to implement custom storage backends for CMF artifacts. This guide shows you how to create Python classes and register them to handle artifact pulls efficiently.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Handle Edge Cases When Artifacts Share URIs but Have Different Names in CMF](/hewlettpackard/cmf/how-to-handle-edge-cases-when-artifacts-share-uris-but-have-different-names)

Learn to handle CMF artifacts sharing URIs but having different names to prevent lineage errors and metadata drift. Understand unexpected framework behavior and resolve potential conflicts.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Use DVC Ingest to Convert Existing Pipelines to CMF](/hewlettpackard/cmf/how-to-use-dvc-ingest-to-convert-existing-pipelines-to-cmf)

Convert existing DVC pipelines to CMF using cmf dvc ingest. Migrate dvc.lock stages to MLMD executions with versioned artifacts for seamless tracking.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Implement Cross-Framework Model Logging (PyTorch & scikit-learn) with CMF](/hewlettpackard/cmf/how-to-implement-cross-framework-model-logging-pytorch-sklearn)

Implement cross-framework model logging for PyTorch and scikit-learn using CMF. Unified metadata tracking for ML models without framework-specific logic. Learn more!

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Migrate from SQLite to PostgreSQL for Production CMF Deployments](/hewlettpackard/cmf/how-to-migrate-from-sqlite-to-postgresql-for-production-cmf-deployments)

Migrate CMF deployments from SQLite to PostgreSQL for production. Configure five POSTGRES environment variables to switch to a scalable PostgreSQL backend.

- Tags: migration-guide
- Published: 2026-03-03

### [Data Slice Tracking in CMF: Analyzing Dataset Subsets with Full Provenance](/hewlettpackard/cmf/how-does-data-slice-tracking-aid-in-analyzing-dataset-subsets)

Analyze dataset subsets with CMF data slice tracking. Get precise model performance insights on specific data segments with full lineage metadata and versioned provenance artifacts.

- Tags: deep-dive
- Published: 2026-03-03

### [How to Debug "DVC not Configured Correctly" Initialization Errors in CMF](/hewlettpackard/cmf/how-to-debug-dvc-not-configured-correctly-initialization-errors)

Debug DVC not configured correctly initialization errors in HPE CMF. Learn how to fix DVC remote configuration issues and prevent metadata tracking failures. Get your CMF project running smoothly.

- Tags: how-to-guide
- Published: 2026-03-03

### [How Execution Reuse Works with `create_new_execution` in CMF](/hewlettpackard/cmf/how-does-execution-reuse-work-with-create-new-execution-parameter)

Learn how CMF execution reuse works with create_new_execution. Set to False to reuse MLMD executions by name or True to create new ones.

- Tags: deep-dive
- Published: 2026-03-03

### [How to Export CMF Lineage Data for External Visualization Tools](/hewlettpackard/cmf/how-to-export-lineage-data-for-external-visualization-tools)

Export CMF lineage data using the CLI or REST API. Generate JSON files or query D3 compatible graphs for Neo4j, GraphViz, and custom visualization tools.

- Tags: how-to-guide
- Published: 2026-03-03

### [How Artifact Version Tracking Prevents Duplicate Dataset Logging in CMF](/hewlettpackard/cmf/how-does-artifact-version-tracking-prevent-duplicate-dataset-logging)

Learn how CMF artifact version tracking prevents duplicate dataset logging using content-derived hashes and ML Metadata to ensure data integrity. Avoid redundant logging effectively.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Implement Fine-Grained Metric Logging per Training Step with CMF](/hewlettpackard/cmf/how-to-implement-fine-grained-metric-logging-per-training-step)

Log fine grained metrics per training step with CMF. Accumulate data in memory and commit to Parquet files with provenance.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Configure and Run the CMF Server with a PostgreSQL Backend](/hewlettpackard/cmf/how-to-configure-and-run-cmf-server-with-postgresql-backend)

Configure and run the CMF server with a PostgreSQL backend by setting environment variables and using docker compose. Learn how to ensure database health checks before CMF initialization.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Use CMF CLI Commands for Metadata Push/Pull Operations](/hewlettpackard/cmf/how-to-use-cmf-cli-commands-for-metadata-push-pull-operations)

Master CMF CLI commands cmf metadata push and cmf metadata pull to sync ML metadata between your local machine and CMF server. Upload execution records and download pipeline states with ease.

- Tags: how-to-guide
- Published: 2026-03-03

### [How Python Environment Tracking Ensures Reproducibility in the CMF Library](/hewlettpackard/cmf/how-does-python-environment-tracking-ensure-reproducibility)

Learn how Python environment tracking in the CMF library captures package specs, versions them with DVC, and links to Git commits to guarantee experiment reproducibility and enable exact reconstruction.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Handle Metadata Merge Conflicts in Multi-Site CMF Deployments](/hewlettpackard/cmf/how-to-handle-metadata-merge-conflicts-in-multi-site-cmf-deployments)

Learn how CMF automatically resolves metadata merge conflicts in multi-site deployments, preventing errors and ensuring idempotent operations for seamless synchronization.

- Tags: how-to-guide
- Published: 2026-03-03

### [How Neo4j Graph Visualization Tracks Pipeline Lineage and Dependencies in CMF](/hewlettpackard/cmf/how-does-neo4j-graph-visualization-track-pipeline-lineage-and-dependencies)

Discover how Neo4j graph visualization tracks pipeline lineage and dependencies in CMF. See live ML pipeline operations and lineage in real-time. Learn more now.

- Tags: deep-dive
- Published: 2026-03-03

### [How to Set Up SSH Remote Storage for Distributed ML Pipelines in CMF](/hewlettpackard/cmf/how-to-set-up-ssh-remote-storage-for-distributed-ml-pipelines)

Set up SSH remote storage for distributed ML pipelines in CMF using cmf init sshremote. Configure DVC and Git for artifact versioning across nodes easily.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Query Tracked Metadata Programmatically Using the CmfQuery API](/hewlettpackard/cmf/how-to-query-tracked-metadata-programmatically-using-cmfquery-api)

Query tracked metadata programmatically with the CmfQuery API. Connect to an MLMD backend and retrieve pipeline, execution, and artifact metadata using Python.

- Tags: how-to-guide
- Published: 2026-03-03

### [How to Implement Custom Auto-Logging for ML Frameworks in CMF](/hewlettpackard/cmf/how-to-implement-custom-auto-logging-for-ml-frameworks-in-cmf)

Implement custom auto logging for ML frameworks in CMF using the step decorator and artifact hierarchy. Effortlessly capture execution lineage for your Python functions.

- Tags: how-to-guide
- Published: 2026-03-03