Train Model

Train a new Digital Twin model from data.

Trains a digital twin from a frame and emits the version it produced.

Durable and long-running — a large panel trains for hours — so its start-to-close ceiling is set by the longest honest run rather than a minute-scale default. experimentMode trains against baseDigitalTwinId under experimentLabel without moving the twin's live version.

At a glance

Module type	TrainModel
Category	Intelligence
Default tier	durable
Runs on	reactive, durable
Starts a flow	no
Holds a body	no
Keeps state between runs	no
Available in	every application type

Parameters

digitalTwinId

string · optional

The twin to train.

digitalTwinName

string · optional

Its name when it was chosen, for the node's label.

digitalTwinVersionId

string · optional

The version this module last produced.

dataViewId

string · optional

Train from a saved data view rather than the data port.

trainConfig

Record<string, unknown> · optional

Training options passed through to the pipeline.

experimentMode

boolean · optional

Train as an experiment off baseDigitalTwinId without moving the twin's live version.

baseDigitalTwinId

string · optional

The twin an experiment branches from.

experimentLabel

string · optional

What to call the experiment.

Inputs

Port	Kind	Required	Description
on	event	no	Implicit on every module that does not itself start a flow: wire an event here to make this module the start of what that event runs.
data	dataframe	yes	The rows to train on.

Outputs

Port	Kind	Connects to	Description
model	json	json, scalar, signal	The twin version this run produced, resolvable by anything downstream.
done	signal	signal	Fires once training has committed.

---

← All application modules