Houdini 20.5 Nodes Geometry nodes

ML Regression Inference geometry node

Apply a model trained using ML Regression Train in a geometry network.

On this page

Overview

This node provides an easy way to apply an ML model that has been trained using ML Train Regression. To use ML Regression Inference, you need to specify the model file and each of the input and output contributions.

Each contribution of both the input and the output can be a point attribute or a volume primitive. The input contributions should match the input contributions specified on the multiparm of ML Example Output. The output contributions should match the target contributions specified on the multiparm of ML Example Output. The tuple sizes and resolutions of the point attributes and volumes must match those sent into the input of ML Example Output.

See Machine Learning documentation for more general information.

Parameters

Batch

In Single mode, a single input is evaluated. In Multiple Packed mode, the model is evaluated on each embedded geometry of the input, the results are stored in corresponding packed primitives.

Model File

The model file that was trained using ML Train Regression.

Reload Model

Force a reload of the .onnx file.

Number of Inputs

Type

Type of input contribution: either a point attribute or a volume.

Point Attribute

Name of a point float attribute.

Volume Name

The name of a volume.

Volume Resolution

#id inputvolumeresolution# Resolution of the volume.

Tuple Size

Tuple size of the point attribute or volume.

Number of Outputs

Type

Type of output contribution: either a point attribute or a volume.

Point Attribute

Name of a point float attribute.

Volume Name

The name of the volume.

Volume Resolution

#id outputvolumeresolution# Resolution of the volume.

Tuple Size

Tuple size of the point attribute or volume.

Inputs

Input Component

The query input component.

Outputs

Output Component

The output predicted by the ML model for the query input component.

Geometry nodes