Skip to main content

Define a loop

Data Intelligence → Loop → New Loop.

  1. Name the loop.
  2. Choose classification or regression.
  3. Add the input fields. Each field has a name and a type: string, integer, float, boolean, array, or object.
  4. Add exactly one output field. That name is the key you will read from a prediction and the key you will send back on train.

Save. The page URL is /loop/{id}. That id is x-loop-id on later calls.

The loop does not learn from predict calls. When you know the outcome, POST it to /train/. Until that call succeeds, the model is unchanged.

Next​