Closed rafaelvalle closed 7 years ago
Thank you for reporting this. I cannot reproduce this locally with fresh conda env, but it looks like apparently an encoding issue. Could you try to clone the repository and apply the following patch and see how it works?
diff --git a/setup.py b/setup.py
index a3f83fe..c5fbc9e 100644
--- a/setup.py
+++ b/setup.py
@@ -107,7 +107,7 @@ if not exists('README.rst'):
create_readme_rst()
if exists('README.rst'):
- README = open('README.rst').read()
+ README = open('README.rst', 'rb').read().decode("utf-8")
else:
README = ''
Yes, indeed the problem is with the ASCII UTF-8 encoding. Another possible solution is to open the file with the UTF-8 encoding.
README = open('README.rst', 'rb', encoding='utf-8').read()
Note that the tacotron_pytorch repo will produce a similar error in train.py for classes TextDataSource and _NPYDataSource
The library supports python2.7, so that solution isn't enough actually. I will try to address same issues in tactoron_pytorch soon.
https://github.com/r9y9/tacotron_pytorch/commit/bdad19fdff22016c7457a979707655bb7a605cd8
Could you confirm if it fixes your issue?
Hi! I get a UnicodeDecodeError when trying to install your library.
Any thoughts what this could be and how to fix it?