yuanke
yuanke
it looks you are hard coding it and limit the vocabulary on lattin letter
I had ran it on gtx1080 about 4days
Am I right about the following items: --dataset more bigger, better performance? --hiddenSize more bigger, better performance? --maxEpoch more bigger, better performance? --batchSize more bigger, better performance?
python3 Python 3.5.2 (default, Nov 17 2016, 17:05:23) [GCC 5.4.0 20160609] on linux Type "help", "copyright", "credits" or "license" for more information. >>> import torch >>> torch.__version__ '0.2.0+8262920' mldl@mldlUB1604:~/ub16_prj/AoAReader$ python3...