Batch Norm Training=True

Screenshot 2022-10-14 091515
What is the need for adding the training=True at the end?? If this wasn’t done how will the training be effected by that?

Which assignment is this from?

Course 4 week 2 resnets.

The exercise instructions mention that “training = training” is used so that training can be enabled or disabled when the function is called. It would be set to True when we use the function for training, and set to False when we use the function for making predictions. This prevents the prediction process from accidentally updating the weights.

If this parameter isn’t given, will it cause any effect in prediction or result?

I believe you can discover the default value of that parameter by looking at the Keras documentation: