search

albertbup / deep-belief-network

A Python implementation of Deep Belief Networks built upon NumPy and TensorFlow with scikit-learn compatibility
MIT License
481 stars 212 forks source link

Hidden units number_ deep-belief-network for regression #34

Closed naloui1 closed 6 years ago

naloui1 commented 6 years ago

Hi, thanks for really good code! I would like to use the deep-belief-network for regression. But, when I see your code, I can see that the number of hidden layers (hidden_layers_structure), the learning rates (learning_rate_rbm, learning_rate), epochs'number (n_epochs_rbm, n_iter_backprop), batch size( batch_size) and activation function are settable. However, this is not the case of the number of hidden units. How is it set please? Thank you Best regards

albertbup commented 6 years ago

Hi, thanks for your comments.

Let me be practical: say hidden_layers_structure=[256, 128], what that means is that your network will have 2 hidden layers, the first one with 256 neurons and the second one with 128. That's the way how you can control the number of hidden units.

Best regards

mahagnyana commented 5 years ago

how about the visible unit albert?