New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Where to get all models as one archive? #26
Comments
Try: This index include all the files. If you do not want to deal with the file, use polyglot download subcommand |
Downloader from pypi uses google and doesn't work at all. Latest version from github shows that everything is downloaded(downloader.download("embeddings2.ru")), but during execution, for example some POS or NER tagging code, shows errors like unexpected EOF of compressed file. Seems that problem in that mirror http://whoisbigger.com/polyglot. When I try to download files manually like embeddings2 ru it just freezes and shows download speed 0 bps. One more problem that by default downloader downloads everything into / directory so after installation I have to go to configuration and set different location. It is better to use some /usr/local/ folder.. |
Yes, pypi package uses the old mirror, you need to use the updated github code. Embeddings are not enough to run POS and NER, you need to download POS and NER models for Russian. The new mirror works, you just did not download all the necessary models. |
Ok I'll try. But is there any command to download everything? Or everything for a single language? |
Or download on demand during code execution? |
Look at the documentation for commands to download all models or all models On Thu, Oct 1, 2015, 12:51 Hodza Nassredin notifications@github.com wrote:
|
My miss. Thanks I found this section. It is hidden in words. Probably better idea is to highjlight it somehow. |
And yes seems that pypi version has correct download directory inside home dir. Probably one more my miss during installation of latest version. Thanks for your work and support. |
Started download for LANG:ru half an hour ago and it is still downloading first file. |
If it is downloading, then it is ok. If you do not see progress in the The new mirror could be slower than the previous one (Google) but it is for On Thu, Oct 1, 2015 at 1:21 PM Hodza Nassredin notifications@github.com
|
OK I'll leave it for night. Hope it will do the thing. |
try to start in parallel another download and see if the other is faster, On Thu, Oct 1, 2015 at 1:25 PM Hodza Nassredin notifications@github.com
|
d) Download l) List u) Update c) Config h) Help q) QuitDownloader> d Download which package (l=list; x=cancel)? d) Download l) List u) Update c) Config h) Help q) QuitDownloader> q |
Make sure that you cleaned your models directory from previously downloaded On Thu, Oct 1, 2015 at 4:17 PM Hodza Nassredin notifications@github.com
|
No. Reason for that error is a corrupted file. hodza@py-trainer:~$ tar -vxjf /home/hodza/polyglot_data/embeddings2/ru/embeddings_pkl.tar.bz2 > /dev/null bzip2: Compressed file ends unexpectedly; It is possible that the compressed file(s) have become corrupted. You can use the `bzip2recover' program to attempt to recover tar: Unexpected EOF in archive |
I\m trying to download models from http://whoisbigger.com/polyglot. But unfortunately it shows 0 bps after some time. Could you give me a link to an alternative donwload?
The text was updated successfully, but these errors were encountered: