Norman Heckscher

Need to refactor with state_is_tuple=True for TF 0.11? ``` cell = cell_fn(args.rnn_size, state_is_tuple=True) self.cell = cell = rnn_cell.MultiRNNCell([cell] * args.num_layers, state_is_tuple=True) ``` ``` feed = {model.input_data: x, model.targets: y} for...

Looking for a way to determine performance. Perhaps it's possible to use perplexity? Something similar to the following pull request: https://github.com/sherjilozair/char-rnn-tensorflow/pull/56

Decay learning rate

1

New decay learning rate based on the following code: https://github.com/sherjilozair/char-rnn-tensorflow/pull/70

Refactor for tensorflow1.0-rc1

1

I realise you are not yet setup to accept PRs. I've signed the Google CLA. Here it is anyway. Cheers.