Model repository
A model repository is a storage system containing machine-learning models and associated versions, artifacts, metadata, or configuration.
Delivering Machine Learning Projects: From design to deployment
A model repository is a managed storage system for machine learning models and their associated versions or metadata.
- How it is used
- Supports controlled, iterative model development and deployment by retaining model instances and their decision history.
- How deep it goes
- Specifies commissioning tasks and record contents, but does not prescribe a repository implementation.
- Examples
- Uses delivery tickets, an experiment-tracking run table, and a repository file-layout figure.
- What you can do after
- Operate/configure it by commissioning a repository and recording model identities, dependencies, inputs, and results for later selection and investigation.
Read how Delivering Machine Learning Projects: From design to deployment explains model repository →
Designing Deep Learning Systems: A guide for software engineers
A model repository is a filesystem-based collection of directories and files that stores deployable model artifacts.
- How it is used
- Organizes deployable model artifacts for discovery, version selection, loading, and inference serving.
- How deep it goes
- Defines filesystem storage and follows discovery, version policy, loading, scheduling, and backend execution without detailing repository synchronization.
- Examples
- Uses Triton model directories and TensorFlow SavedModel layouts with versioned subdirectories, assets, variables, and saved_model.pb.
- What you can do after
- Operate/configure TensorFlow Serving or Triton around filesystem-hosted model artifacts and selected model versions.
Read how Designing Deep Learning Systems: A guide for software engineers explains model repository →
Related concepts
- NVIDIA Triton Inference Serviceused by · NVIDIA Triton Inference Server is open-source software for deploying and serving machine-learning models through standardized interfaces across frameworks and hardware backends.
- language modelhas part · A language model is a computational model that represents patterns in linguistic tokens to estimate sequence probabilities and predict or generate language from context.
- versionused by · A version is a distinct, identifiable state or revision of an artifact, system, or product.
- foundational modelhas part · A base or foundation model is a pretrained model that provides learned parameters or capabilities as a starting point for reuse or adaptation to downstream tasks.
- metadatahas part · Metadata is information that describes, identifies, or provides context about other data, resources, or artifacts.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.