2

I'm trying to use Syntaxnet pretrained models for Italian language. I just found one issue related to tokenization task on Italian and Spanish languages. I have got a preliminar problem. The posted link for pretrained dataset dowload does not work. Here it is : https://github.com/tensorflow/models/blob/master/syntaxnet/universal.md. Could someone provide me a working repository link? Thanks a lot

fiamminaM
  • 21
  • 2

1 Answers1

1

Here it is: http://download.tensorflow.org/models/parsey_universal/Italian.zip

By the way, you can use this project for Docker as a basis for your own container. Just replace the model for Russian with the model for Italian. https://github.com/IINemo/docker-syntaxnet_rus

This container implements standalone server that can be used in conjunction with syntaxnet wrapper: https://github.com/IINemo/syntaxnet_wrapper

Chen Gupta
  • 131
  • 2
  • 12
  • thanks. I just fixed the broken link in order to access to Italian Language Pretrained Models. thanks @ Chen Gupta – fiamminaM May 02 '17 at 10:16
  • By the way, you can generate the SyntaxNet docker container for Italian with this [project](https://github.com/IINemo/docker-syntaxnet) now using the command from docker/main: `./build.sh it_IT.UTF-8 http://download.tensorflow.org/models/parsey_universal/Italian.zip` Or just run: `echo "La bellezza ha una verità tutta sua " | docker run --rm -i inemo/syntaxnet_ita` – Chen Gupta May 03 '17 at 09:24