Build with Causilo.

Predict new rows from the examples in your table. No training step required.

Make your first call

Client callPYTHON

prediction, metadata = cx.predict_with_metadata(
    context,
    query,
    target="renewed",
    model="causilo-clf",
    output_type="preds",
)

print(prediction)
print(metadata["quota"])

Client-call excerpt. Use an existing API key and prepare tables in Quickstart.

Causilo models

One interface for classification and regression. Choose the model for the target you want to infer.

Classificationcausilo-clfInfer class labels or class probabilities from the examples in your context.Regressioncausilo-regInfer numeric values, including means, medians and quantiles.

Work with your data

The context includes known answers in the target column. The query carries the same feature columns, in the same order, without the target.

Context, query and targetChoose a target and prepare compatible tables.Runnable examplesExplore complete scripts and their recorded stdout.

Check the cost before uploading

Cells count context and query rows together, excluding the target column. Stored contexts save the upload, not the cells. Read credential usage and allowances with GET /v1/usage; check request ceilings with GET /v1/limits.

Limits and usageCheck your table shape and understand what counts.Resolve an errorUse the error code and retryable flag to choose the next step.

API reference

Request and responseFind routes, parameters and response fields.

Read the Terms of Service and Privacy Policy. Questions: api@nums.world.