A series of Jupyter notebooks that walk you through the fundamentals of Machine Learning and Deep Learning in Python using Scikit-Learn, Keras and TensorFlow 2.
Apache License 2.0
7.44k
stars
3k
forks
source link
Fix Model Fit Error for : Compiling Functional API with Aux Output with one loss function #127
Model compile was throwing error for Functional API with aux output.For below code of Chapter 10 Pg 334 was throwing error
while model training on fit method call.
model.compile(loss=("mse", "mse"), loss_weights=(0.9, 0.1), optimizer=optimizer, metrics=["RootMeanSquaredError"])
ValueError: For a model with multiple outputs, when providing the metrics argument as a list, it should have as many entries as the model has outputs. Received:
metrics=['RootMeanSquaredError']
of length 1 whereas the model has 2 outputs.
We need to specify another loss function for aux output like for fit method to work correctly.
model.compile(loss=("mse", "mse"), loss_weights=(0.9, 0.1), optimizer=optimizer, metrics=["RootMeanSquaredError", "RootMeanSquaredError"])
Model compile was throwing error for Functional API with aux output.For below code of Chapter 10 Pg 334 was throwing error while model training on fit method call. model.compile(loss=("mse", "mse"), loss_weights=(0.9, 0.1), optimizer=optimizer, metrics=["RootMeanSquaredError"])
ValueError: For a model with multiple outputs, when providing the
metrics
argument as a list, it should have as many entries as the model has outputs. Received: metrics=['RootMeanSquaredError'] of length 1 whereas the model has 2 outputs.We need to specify another loss function for aux output like for fit method to work correctly. model.compile(loss=("mse", "mse"), loss_weights=(0.9, 0.1), optimizer=optimizer, metrics=["RootMeanSquaredError", "RootMeanSquaredError"])