hugovk / char-rnn-tensorflow / 44

93%

master: 93%

push

travis-ci

Split dropout into input and output parameters

Usage example:
```
python train.py --input_keep_prob=0.8 --output_keep_prob=0.5
```

Useing dropout will likely slow down training and
may not prevent overfitting. For more details read:
https://www.cs.toronto.edu/~hinton/absps/JMLRdropout.pdf

Any code review here would be welcome!

5 of 5 new or added lines in 2 files covered. (100.0%)

242 of 261 relevant lines covered (92.72%)

1.85 hits per line