-1

I was just wondering if anyone has a pre-trained model for syntaxnet. We are trying to save as much time as possible and this would be really helpful.

Cheers.

Matt
  • 71
  • 1
  • 4

2 Answers2

0

As far as I know SyntaxNet is already trained with Penn Treebank. but if you want to retrain it with other training set, for example Universal Dependencies, here is a useful link which has the training .conll data set, how to convert it to a readable format for SyntaxNet training script, and how to train it the dependency parsing and Part of Speech tagger.

Nazanin Tajik
  • 412
  • 2
  • 15
0

In case of interest in a pretrained model for some language rather than English:

According to that update from Google, the company has just released pretrained syntactic models for 40 languages (and named these models as Parsey McParseface's cousins).

Instructions on how these models could be downloaded, along with evaluation results achieved by them, are available on the following GitHub's page.

upokatik
  • 13
  • 4