Waffles currently does not support removing a layer in a neural network. Its not too difficult to make this work, the only issue is that beginIncrementalLearning ( or other such setup method ) would need to be called again after a layer was removed in order to set up the weights and numbers of neurons. This can definitely lose data, but it could be useful to simply remove a layer, rather than copy the weights we want into a new neural network.
Waffles currently does not support removing a layer in a neural network. Its not too difficult to make this work, the only issue is that beginIncrementalLearning ( or other such setup method ) would need to be called again after a layer was removed in order to set up the weights and numbers of neurons. This can definitely lose data, but it could be useful to simply remove a layer, rather than copy the weights we want into a new neural network.