Partial Least Squares Outputs
The following outputs can be generated by the Partial Least Squares tool. Depending on your configuration and analysis requirements, you can select one or more outputs from the Outputs section.
Ouptut | Type | Description |
|---|---|---|
Predicted Target | Signal | The value predicted by the Partial Least Squares model for the target variable based on the input signals. |
Coefficients | Scalar | Shows the influence of each input signal on the predicted target. Larger coefficient magnitudes indicate a stronger impact on the prediction. |
Intercept | Scalar | The baseline prediction value used by the model before the influence of the input signals is applied. |
Model input properties | Scalar | Provides information about the model inputs used during training. This output can be used for model validation and troubleshooting. |
R squared | Scalar | Indicates how well the model fits the training data. Values closer to 1 indicate that the model explains a larger portion of the variation in the target variable. |
Training window | Condition | Identifies the time range used to train the Partial Least Squares model. This output can be used to verify the training period and compare it with scored data. |
X loadings | Scalar | Shows how the input signals contribute to the latent variables identified by the model. Useful for understanding relationships between input signals. |
X weights | Scalar | Shows the importance of each input signal when constructing the latent variables used by the model. |
Y loadings | Scalar | Shows how the target variable relates to the latent variables identified by the model. |
Y weights | Scalar | Shows the contribution of the target variable to the latent variables used by the model. |