chainer-char-rnn
chainer-char-rnn copied to clipboard
yusuketomoto
karpathy's char-rnn (https://github.com/karpathy/char-rnn) implementation by Chainer
Upon running `python train.py`, I receive this: ``` data/tinyshakespeare/input.txt corpus length: 1115394 vocab size: 65 Traceback (most recent call last): File "train.py", line 67, in cuda.get_device(args.gpu).use() File "cupy/cuda/device.pyx", line 73,...
I tried to upgrade to use chainer v3 and cupy v2. If you have a time, could you take a look? Thanks for the great implementation. I appreciate for your...
thank you for implementing it in chainer. I've ported your code to chainer v2 and python 3.6. https://github.com/ranjita-naik/Deep-Learning/tree/master/RNN
Is there a way to set the number of layers, if not what is the default?
Chainer with Multi CPU Support for Linux (Chainer-CHAR-RNN Speed improvements) wrote little perl script to do it changed sample.py file so that you dont need to give these arguments to...
Hello, Another question. I know char-rnn has started all this. Are there any comparisons out there which compares the accuracy/speed of char-rnn and chainer implementation. Getting Torch installed is a...
Hello, Is there anyone out who used this library to train a text dataset using word level encodings instead of characters? What are your thoughts, in terms of training speed...
Hi, Thank you for the great contribution. The program works fine with tinyshakespeare dataset and other dataset, however part of "train.py" code looks quite strange to me. Line 87-91: for...
Hi, Thanks a lot for your example of char_rnn. I have a question about your usage of unchain_backward (line 107) in the train.py. Assume batchsize is 1, and the code...
If gpu argument is set to 0, it still attempts to run with CUDA on as the test was for >= rather than just >. To get this code to...
Metadata
Owner
Metadata
karpathy's char-rnn (https://github.com/karpathy/char-rnn) implementation by Chainer