Body
JSON request structures for various models
The request body should be passed using the -d option in the cURL command. Replace [data] with the actual JSON-formatted content you want to send for prediction. The structure and required fields depend on the specific model’s configuration. (See the next section for details.)
The v1 request body uses a structured format built from nested arrays of objects. Each object represents a feature (column) and contains two keys:
"alias": the name of the input column as defined during model training
"selectedValue": the value you are submitting for prediction
This format is often used when data is being prepared manually or when values are mapped from UI elements.
v1 Example - Binary Classification / Multiclass Classification / Regression)
{
"data": {
"predict_values": [
[
{ "alias": "Lead Source", "selectedValue": "bing" },
{ "alias": "Lead Origin", "selectedValue": "API" },
{ "alias": "Converted", "selectedValue": "YES" }
],
[
{ "alias": "Lead Source", "selectedValue": "Google" },
{ "alias": "Lead Origin", "selectedValue": "Landing Page Submission" },
{ "alias": "Converted", "selectedValue": "YES" }
],
[
{ "alias": "Lead Source", "selectedValue": "Olark Chat" },
{ "alias": "Lead Origin", "selectedValue": "API" },
{ "alias": "Converted", "selectedValue": "YES" }
]
]
}
}
v1 Example (Timeseries)
{
"data": {
"predict_values": {
"startDate": "2023-04-17",
"endDate": "2023-04-18",
"sequenceID": "A"
}
}
}
For time series forecasting, the structure differs from classification and regression models.
startDate and endDate: define the prediction period
sequenceID: identifies the specific time series to forecast (e.g., a product, store, or region)
JSON structure represents a data object with a key "data" mapping to an object called "predict_values". The "predict_values" object contains three key-value pairs. The required keys are "startDate", "endDate", and "sequenceID", and their corresponding values are "2025-04-17", "2025-04-18", and "A" respectively.
Last updated
Was this helpful?