summarize_model

summarize_model(model, tflite=False, build=False, test=False, built_model=None)

Generate a summary of the given model and return the summary as a string

See also

• Model Summary Guide

• Model summary API examples

Parameters:

• model (Union[str, MltkModel, Model, TfliteModel]) –

  Either

  • a path to a .tflite, .h5, .mltk.zip, .py file,

  • or a mltk.core.MltkModel, mltk.core.KerasModel,

  • or mltk.core.TfliteModel instance

• tflite (bool) – If true, the return the summary of the corresponding .tflite model. If true and model= mltk.core.KerasModel, this will quantize it into a .tflite model

• build (bool) – If true, then generate a .h5 or .tflite by training the given mltk.core.MltkModel model for 1 epoch. This is useful for summarizing the mltk.core.MltkModel without fully training the model first

• test (bool) – If true and the model is the name of a MltkModel, then load the MltkModel in testing mode

• built_model (Union[Model, TfliteModel]) – Optional, previously built mltk.core.KerasModel or mltk.core.TfliteModel associated with given mltk.core.MltkModel

Return type:

str

Returns:

A summary of the given model as a string