-
https://arxiv.org/abs/1709.00028
-
Using `1.6.0` (bda9b18b6cede63b0476e9c144da3f62f03515b1), with `--type transformer`,
it's looking for 3 inputs instead of 2.
With https://gist.github.com/emjotde/9c5260870c25304b9c8b111ddcf81b74…
-
I made a fresh install on a CentOS server, with python 2.7.13 (anaconda) and TF 1.1. python, TF and g2p-seq2seq are installed in userspace (with the --user flag)
Running python2 setup.py test give…
-
Thanks for sharing your code which is pretty clear and not as bloated as the other repositories showcasing sequence-to-sequence networks with attention.
Unfortunately the dimensions do not add up a…
-
when i run
`python train.py --usegpu --nepochs 10 --batchsize 1 --nworkers 1`
I got the error below. I am in China. I wonder if there is the internet block. And what should I do?
Architecture(…
-
Hi,
I try to train a model using a custom 16kHz database. Tacotron 1 model training is successfully finished. When it comes to the wavenet training, i got the following error:
[Condition x == y di…
-
Hi everyone,
I'm implementing a model with PyTorch and realised when I call my GRU forward method, I raise an exception (due to bad shape) but there is an error in the message given due to undefine…
-
@ines @honnibal
Python version: 2.7.6
Platform: Linux-3.16.0-77-generic-x86_64-with-Ubuntu-14.04-trusty
spaCy version: 2.0.0a17
Models: en, en_core_web_sm, xx_ent_wiki_sm
def …
-
Hi,
Great repo! :-)
I have a question about the pretrained models. If I create a model with the indicated options (rnn_hidden_size=800, nb_layers=5, rnn_type= 'gru',bidirectional=False) I get …
-
Hi @taolei87 ,
I have a question about weight matrix dimension,
In the SRUCell code, I found the ` k = 4 if n_in != out_size else 3`
But When I read the paper, it's only have 3 weight matrix, W,…