Hello, My name is Azis Im from Indonesia. I had a question, why when i use one layer for convolusional and pooling layer the training time only runs for a short time and only does training in a few epochs even though i don’t use callback
There are no trainable parameters in a pooling layer, so the only computational work to do is in your one Convolutional layer. That would account for speedy execution of each iteration. The number of training epochs is established in the model.fit() function. If you don’t specify a number, it defaults to 1. How many epochs did you specify? How many were executed?
i use 15 epochs and only one which executed
I am not aware of any hidden or automated early stopping for a training cycle that runs properly ie not throwing any exceptions.
Since the default is 1 are you sure the epochs=n parameter is spelled and passed correctly to model.fit() ?
yes i passed correctly
Can you share the notebook you are using in pdf form so that we can have a better look at the issue
Thanks