Error: Build Error Using TimeDistributed Recurrent Layer with Dropout

[sjebbara]

Hi,
I was building a network that processes sequences of sequences (batch_size, n_sequences, n_elements, element_size) with a TimeDistributed RNN layer. The following Gist shows a minimal example:
https://gist.github.com/sjebbara/620d9ac4fc389a3f454444ec76c764f3

The odd thing is that the network compiles if dropout_W and dropout_U of the RNN (here GRU) are set to 0. As soon as one of them is greater than 0 the models model._make_predict_function() throws an error with the message:

theano.gof.fg.MissingInputError:
An input of the graph, used to compute Subtensor{::, int64, ::}(<TensorType(float32, 3D)>, Constant{0}), was not provided and not given a value.Use the Theano flag exception_verbosity='high',for more information on this error.`

The error emerges here:

...
File "/homes/sjebbara/git/keras-original/keras/engine/topology.py", line 514, in call
self.add_inbound_node(inbound_layers, node_indices, tensor_indices)
File "/homes/sjebbara/git/keras-original/keras/engine/topology.py", line 572, in add_inbound_node
Node.create_node(self, inbound_layers, node_indices, tensor_indices)
File "/homes/sjebbara/git/keras-original/keras/engine/topology.py", line 149, in create_node
output_tensors = to_list(outbound_layer.call(input_tensors[0], mask=input_masks[0]))
File "/homes/sjebbara/git/keras-original/keras/layers/wrappers.py", line 120, in call
initial_states=[])
File "/homes/sjebbara/git/keras-original/keras/backend/theano_backend.py", line 887, in rnn
go_backwards=go_backwards)

Since it only arises when dropout is applied I suspect the error to originate from the get_constants method in the GRU.

Please make sure that the boxes below are checked before you submit your issue. Thank you!

• Check that you are up-to-date with the master branch of Keras. You can update with:
  pip install git+git://github.com/fchollet/keras.git --upgrade --no-deps
• If running on Theano, check that you are up-to-date with the master branch of Theano. You can update with:
  pip install git+git://github.com/Theano/Theano.git --upgrade --no-deps
• Provide a link to a GitHub Gist of a Python script that can reproduce your issue (or just copy the script here if it is short).