Oliver Hennigh

This worked for me! (tensorflow 1.0 alpha

That is correct, the flow conditions are not input to the network but they are kept consistent through the dataset. I have since played around with adding flow conditions with...

This was caused by an update to Tensorflow but I changed it should be working now.

That is correct, I was doing a division and I was expecting an int but it gave a float because of python 3. I just updated this and it should...

Ya for sure, glad it was helpful. Hope it works well for you. On Oct 6, 2016 4:25 AM, "Nick Shah" [email protected] wrote: > Hey Oliver, > > Great job...

Thanks for the comment! The hidden state is reset to zero after every batch automatically. This is because in the computational graph the first hidden state is a tf.zero node...

hmm, Im not really sure whats happening with just this info. I think it might be that you need to specify the batch size or you are giving it the...

That is what is happening with the network architecture. If you want a better architecture consider checking out the branch here https://github.com/loliverhennigh/Convolutional-LSTM-in-Tensorflow/tree/residual_model . It uses residual connections and stuff. I...