Skip to content

This section describes Downstream Model interface and configuration in detail.

Once you navigate to Downstream Models after you log into BaseModel, you will see a listing of all models created so far.

Interface Overview: Downstream Models in BaseModel

The Downstream Models interface in BaseModel provides users with a comprehensive view and control over Downstream Models used for model training, scoring, and other tasks. Below is a breakdown of the key components of this interface:

1. Navigation Panel (Left Sidebar)

Downstream Models navigation panel

Purpose:

The left sidebar contains filtering options to help users organize and view specific models.

Filter by Status:

Status Description
Scheduled Models that are planned for training.
Training Models that are currently being trained.
Ready Models that have completed training and are ready for use.
Training Failed Models where training encountered errors.
Draft Models that are still in draft mode.

Filter by Type:

Type
Classification - Binary
Classification - Multiclass
Classification - Multilabel
Recommendation
Regression

2. Downstream Models List (Main Content Area)

Purpose:

Displays a list of all Downstream Models with key details about each.

Key Components:

Column Description
Name The identifier for each Downstream Model.
Type Indicates whether the model is for regression, classification, or recommendation.
Status Displays the current state of the model (e.g., "READY").
Last Run Shows when the model was last executed.
Next Run Displays the next scheduled run, if applicable.
Upstream Indicates the source dataset or model used as an upstream input.
Edited Shows the last modification date.

Additional Features:

  • Search and Filter Results: Displays the total number of results (e.g., "155 Results"), which updates based on applied filters.
  • Model Actions (Three Dots Menu): Each model has an options menu where users can perform actions like editing, deactivating, and quickly viewing the upstream model.