MLOps · Schema

MLOps Model

Schema describing a machine learning model artifact tracked by an MLOps process, including version, framework, training inputs, evaluation metrics, lineage, and governance metadata.

AI OperationsCRISP-ML(Q)DevOpsMachine LearningML EngineeringML GovernanceML PipelinesModel DeploymentModel MonitoringModel Serving

Properties

Name Type Description
id string Unique identifier for the model in the registry.
name string Human-readable name of the model.
version string Semantic or registry-assigned version of the model artifact.
description string
framework string
task string
trainingDatasetId string
metrics object Map of metric name to numeric score (e.g., accuracy, f1, rmse).
owner string
stage string
createdAt string
updatedAt string
tags array
View JSON Schema on GitHub

JSON Schema

mlops-model-schema.json Raw ↑ 
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://raw.githubusercontent.com/api-evangelist/mlops/main/json-schema/mlops-model-schema.json",
  "title": "MLOps Model",
  "description": "Schema describing a machine learning model artifact tracked by an MLOps process, including version, framework, training inputs, evaluation metrics, lineage, and governance metadata.",
  "type": "object",
  "required": ["name", "version", "framework", "task"],
  "properties": {
    "id": {
      "type": "string",
      "description": "Unique identifier for the model in the registry."
    },
    "name": {
      "type": "string",
      "description": "Human-readable name of the model."
    },
    "version": {
      "type": "string",
      "description": "Semantic or registry-assigned version of the model artifact."
    },
    "description": {
      "type": "string"
    },
    "framework": {
      "type": "string",
      "enum": ["tensorflow", "pytorch", "scikit-learn", "xgboost", "lightgbm", "huggingface", "onnx", "other"]
    },
    "task": {
      "type": "string",
      "enum": ["classification", "regression", "ranking", "clustering", "generation", "embedding", "recommendation", "forecasting", "other"]
    },
    "trainingDatasetId": {
      "type": "string"
    },
    "metrics": {
      "type": "object",
      "additionalProperties": { "type": "number" },
      "description": "Map of metric name to numeric score (e.g., accuracy, f1, rmse)."
    },
    "owner": {
      "type": "string"
    },
    "stage": {
      "type": "string",
      "enum": ["development", "staging", "production", "archived"]
    },
    "createdAt": {
      "type": "string",
      "format": "date-time"
    },
    "updatedAt": {
      "type": "string",
      "format": "date-time"
    },
    "tags": {
      "type": "array",
      "items": { "type": "string" }
    }
  }
}