Ehud Ben-Reuven

On Ubuntu (inside libsnark docker) I had to do ``` git submodule update --init --recursive ```

if you are using a different input.txt when re-running the training code with an old model (using -init_from flag) then you need more changes to the code. I have support...

Looks like something wrong with Keras. According to their [manual](https://keras.io/utils/#get_file) the `get_file(path, ...` command should have placed the file at `{datadir}/{path}`

1. just check the code which wrote the history (last cell of train notebook) 2. This message is OK, I decided not to load the entire model but just the...

there was a small bug in code which I now updated. In any case on first run FN1 should be None and no load should be made

It looks like you run out of RAM on your Amazon instance. You can open a terminal and use htop to see this happening while the code is running. As...

try training with nflips=0 it may help

thats ok or change the value assigned to nflips in cell 9 of train.ipynb

I'm using Theano so there could be bugs related to TF which I did not noticed. I found that Theano runs on this problem about 5 times faster than TF...

generating with BiLSTM is tricky. In generation you move forward step by step generating one word at a time. However the BiLSTM model requires you to move both forward and...