Model Input / Output Types

Understanding Model Schemas within Arthur

Role of a Model Schema in Arthur

Arthur's model schema records important properties for your model's attributes, including their value type and stage. These data types define the data that enters and exits your model.

  • The InputType of a model specifies whether data enter your model as a tabular data frame, image, or raw text.
  • The OutputType of a model specifies the modeling task: whether your model predicts values for a regression task, probabilities for a classification task, bounding boxes for a computer vision object-detection task, or token sequences (and their probabilities for generative tasks).

Supported Schemas in Arthur

Arthur supports several input and output types.

The input types supported are:

The output types supported are:

What’s Next

Quick look at model monitoring in Arthur by type