Get Model Configuration

Page Outline

• Authentication Requirements
• Path Parameters
• Operational Behavior
• Response
  • Example JSON Response (Built-in Algorithm)
• Error Codes

---

GET/api/v1/models/{model_id}/config

The Get Model Configuration endpoint fetches the current operative training and pre-processing configuration for a given model architecture. This includes essential details such as the selected algorithmic framework, hyperparameter definitions, feature column assignments, and data imputation strategies. It is meant to provide a deterministic snapshot of how a model will train when the Start Training endpoint is invoked.

[!TIP] The data returned here represents the effective configuration. If certain hyperparameters are omitted during model creation, the API automatically inherits organizational or framework defaults, which will be visible in this response payload.

Authentication Requirements

This endpoint necessitates a secure Bearer token for access.

• Header: Authorization: Bearer <Your-JWT-Token>
• Permissions Required: The authenticated programmatic user must possess minimum model:read or organization:viewer roles.

Path Parameters

Parameter	Type	Required	Description
model_id	string	Yes	The unique UUID identifier (mdl-xxx) of the model whose configuration schema you are inspecting.

Operational Behavior

When requesting the active configuration, understand how the backend synthesizes the response mapping:

1. Effective Configuration Merging: The response reflects an immutable merge between system default fallbacks, organizational overrides, and explicit model-level mutations applied during Update Configuration.
2. Custom Model Specifications: If the model_id targets an entity with model_type: "custom", the payload dynamically injects Custom Model fields (e.g., entry_point, runtime_environment, resource_profile, requirements_file).
3. Audit Trailing Policy: Reading a configuration is considered a safe, non-mutating action and therefore does not trigger an event capture into the system's Audit Logs.

Response

The API responds with an HTTP 200 OK status and returns a comprehensive ModelConfig JSON object detailing the hyperparameter scope.

Example JSON Response (Built-in Algorithm)

{
  "model_id": "mdl-7f3a19bc92",
  "algorithm": "xgboost_classifier",
  "target_column": "is_churned",
  "feature_columns": ["account_age_days", "monthly_mrr", "support_tickets_l30d"],
  "hyperparameters": {
    "n_estimators": 250,
    "max_depth": 7,
    "learning_rate": 0.015,
    "subsample": 0.85,
    "colsample_bytree": 1.0,
    "objective": "binary:logistic"
  },
  "preprocessing": {
    "handle_missing": "median",
    "scale_features": true,
    "encode_categoricals": "target_encoding"
  },
  "effective_from": "org_tuning_profile_v2"
}

Error Codes

Prepare your client to handle the following error states:

Code	Reason	Resolution
401	Unauthorized	Ensure a valid JWT is present in the Authorization header.
403	Forbidden	Ensure your user credentials possess model:read scope.
404	Not Found	Verify the model_id string and organization scope boundary.
429	Too Many Requests	Rate quota exceeded across the organization's tenant.