pytensor_ml.model.Model.predict#

Model.predict(*inputs, **named_inputs)#

Run the inference pass, dropping dropout and reading batch norm’s running statistics.

Parameters:
*inputsndarray

One array per data input, in the order the graph reports them. Positional order follows graph traversal rather than the order the inputs were built in, so name them instead whenever a model reads more than one.

**named_inputsndarray

The same arrays given by the name of the variable each belongs to, which is order-free.

Returns:
predictionsndarray

The model’s output for the given data.