Enqueue a new training task to Fount
| Time | Status | User Agent | |
|---|---|---|---|
Retrieving recent requests… | |||
Enqueue a new training task to Fount.Training progress can be monitored using the returned job_name with the task-status endpoint.
Submit a model training job to Fount. This endpoint enqueues a new training task and runs it using Fount's causal AI engine.
Provide your uploaded dataset ID, define which columns are categorical, which is the date column, and which are your targets. Fount will learn causal relationships across all inputs and outputs and return a job_name you can use to monitor progress via the task-status endpoint.
Authentication: Include your API key in the X-API-KEY request header.
Note: Training runs asynchronously. The response returns immediately with a job_name, use that with the task-status endpoint to poll for completion.
Parameter descriptions
These go on each individual field in the ModelConfiguration request body schema:
| Parameter | Description |
|---|---|
dataset | The ID of the dataset to train on. Returned when you upload a file via the Upload endpoint. |
model_name | A unique name for this trained model. Used to identify and reference the model in inference calls. Choose something descriptive, such as "demand_forecast_q4". |
categorical_cols | List of column names in your dataset that contain categorical values, for example, product category, region, or channel. Pass an empty array if your dataset has no categorical columns. |
date_col | The name of the date or timestamp column in your dataset. Fount uses this to learn time-based causal relationships. Pass null if your dataset has no date column. |
target_cols | List of column names you want Fount to model and forecast, for example, ["Sales", "Revenue"]. Must be numeric columns. At least one target is required. |
validation_data_required | Whether to hold out a portion of the dataset for validation during training. Set to true to enable validation, false to train on the full dataset. |
validation_split | The proportion of the dataset to use for validation when validation_data_required is true. Accepts a value between 0 and 1, for example, 0.2 reserves 20% of the data for validation. Ignored if validation_data_required is false. |
time_granularity | The time interval between rows in your dataset. Tells Fount how to interpret the spacing between data points. Defaults to "daily". |