search

yu4u / convnet-drawer

Python script for illustrating Convolutional Neural Networks (CNN) using Keras-like model definitions
MIT License
596 stars 98 forks source link

I really love your work, however #8

Open elsheikh21 opened 5 years ago

elsheikh21 commented 5 years ago

I really love your work, as I said earlier but why don't you add the other layers, (Batch Normalization, Activation, Dropout, Zero Padding 2D, Max Pooling 2D), or at least package your code?

I am ready and willing to contribute with you, if you want..

yu4u commented 5 years ago

Thank you for your suggestion. Max Pooling 2D is implemented. I think the other layers are somewhat annoying if they are visualized. One solution is to add these information (e.g. activation or bn) to the text under each layer. I had this idea before but not implemented. I'll try later.

elsheikh21 commented 5 years ago

My bad, for mentioning something that was already implemented 😄 Seems like you know that it easy, do it man and package. You are hindering your work this way

elsheikh21 commented 5 years ago

Yes, it is. However, as I said before I can contribute to the project with you if you want

On Thu, Jan 17, 2019 at 7:51 PM Koglin notifications@github.com wrote:

Be aware that you can do BN before activation (e.g. dense, BN, softmax). And that is done regularly since it gains better results. It's kind of sad that the model summary does not incorporate that important information.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/yu4u/convnet-drawer/issues/8#issuecomment-455286350, or mute the thread https://github.com/notifications/unsubscribe-auth/AY207fzesDw7RQbJLfLJh9Mx-ESPClUeks5vEMZHgaJpZM4ZkaQB .