Compatibility with normal multilayer perceptron

pplonski / keras2cpp

This is a bunch of code to port Keras neural network model into pure C++.

MIT License

679 stars 153 forks source link

Compatibility with normal multilayer perceptron #15

Open masakinakada opened 7 years ago

masakinakada commented 7 years ago

Hi @pplonski Thank you so much for this amazing keras converter. I have MLP trained using keras and want to use in C++.

Is this working with MLP which has relu and linear as an activation function? I noticed that you mentioned you focused on CNN.

Thank you very much in advance. Masaki

pplonski commented 7 years ago

Hi! Right now, there is softmax and Relu activation implemented. Adding linear activation should be easy. Would you add it? In function keras::LayerActivation::compute_output

masakinakada commented 7 years ago

Yes I added it! As you said, linear layer was easy. Just wanted to make sure other parts are fine too!

Thanks you @pplonski !

pplonski commented 7 years ago

Great! Would you like to pull your changes?

masakinakada commented 7 years ago

push my change?

pplonski commented 7 years ago

Sure, please make pull request.