Stack Overflow
Stack Overflow

Questions tagged [parsey-mcparseface]

Parsey McParseface is the English parser at the heart of SyntaxNet, a Natural Language Understanding framework for Tensorflow.

Parsey McParseface is the English parser at the heart of Google's SyntaxNet, a Natural Language Understanding (NLU) framework for tensorflow. The name is a reference to Boaty McBoatface, the result of an online poll to name Britain's next Royal Research Ship.

22 questions
14
votes
4 answers

Syntaxnet / Parsey McParseface python API

I've installed syntaxnet and am able to run the parser with the provided demo script. Ideally, I would like to run it directly from python. The only code I found was this: import subprocess import…
zvisofer
  • 1,346
  • 18
  • 41
10
votes
1 answer

Is it possible to export a syntaxnet model (Parsey McParseface) to serve with TensorFlow Serving?

I have the demo.sh working fine and I've looked at the parser_eval.py and grokked it all to some extent. However, I don't see how to serve this model using TensorFlow Serving. There are two issues I can see off the top: 1) There's no exported model…
dmansfield
  • 1,108
  • 10
  • 22
9
votes
3 answers

NLP of Legal Texts?

I have a corpus of a few 100-thousand legal documents (mostly from the European Union) – laws, commentary, court documents etc. I am trying to algorithmically make some sense of them. I have modeled the known relationships (temporal,…
Matthias Winkelmann
  • 15,870
  • 7
  • 64
  • 76
7
votes
1 answer

Definition of POS tag and Dependency label sets are used within Parsey McParseface?

The POS tags and Depedency labels output by Parsey McParseface are given in the tag-set and label-set files here respectively. The Syntaxnet readme outlines that the model was trained on the Penn Treebank, OntoNotes and the English Web Treebanks.…
Michael Thomas
  • 1,354
  • 7
  • 18
6
votes
1 answer

How to get constituency-based parse tree from Parsey McParseface

Parsey McParsey returns a dependency-based parse tree by default, but is their a way to get a constituency-based parse tree from it? EDIT: To clarify, by "to get from it" I mean from the Parsey itself. Though building a tree from ConLL output would…
maga
  • 720
  • 3
  • 13
4
votes
1 answer

How to test SyntaxNet trained model (Spanish UD)?

Training my own SyntaxNet model for Spanish-Ancora UD corpus following the instructions from here did not give me errors. After all the steps the final files it creates were: -category-map -char-map -checkpoint -context -graph -label-map …
Octavio Soto
  • 781
  • 9
  • 13
4
votes
3 answers

Parsey McParseface incorrectly identifying root on questions

It seems to me that Parsey has severe issues with correctly tagging questions and any sentence with "is" in it. Text: Is Barrack Obama from Hawaii? GCloud Tokens (correct): Is - [root] VERB Barrack - [nn] NOUN Obama - [nsubj] NOUN from - [adp]…
Phil
  • 41
  • 2
3
votes
0 answers

Syntaxnet C++ Api

I have installed Syntaxnet and parsed few example files. I am planning to create a C++ application which uses Syntaxnet to parse the input document. I searched to find if I could use the Syntaxnet Net C++ API, but i couldn't find one. Can any one…
H414
  • 71
  • 8
3
votes
2 answers

How to deploy parsey's cousins with tensorflow serving

Are there instructions or some documentation somewhere or could somebody describe how to deploy the models available as "Parsey's Cousins" (see https://github.com/tensorflow/models/blob/master/syntaxnet/universal.md) with SyntaxNet under Tensorflow…
jpp1
  • 2,019
  • 3
  • 22
  • 43
1
vote
0 answers

Problem installing SyntaxNet on Ubuntu 18.04.1: Error executing Jupyter command 'nbextension': [Errno 2] No such file or directory

I want to use Parsey McParseface, and I'm following the installation instructions for SyntaxNet here: https://github.com/tensorflow/models/tree/master/research/syntaxnet#ubuntu-1604-binary-installation#ubuntu-1604-binary-installation I'm using…
Brad Spendlove
  • 43
  • 1
  • 6
1
vote
0 answers

Syntaxnet POS tagger use of capitalization

I want to use Syntaxnet for getting the POS tags of tweets (and more specifically, extracting named entities from the text). However, Parsey McParseface is case-sensitive by default. Since tweets often not use capitalization, I was thinking of using…
Guido
  • 6,182
  • 1
  • 29
  • 50
1
vote
1 answer

Obtain relationship between words of a sentence

I am working on a project based something on natural language understanding. So, what I am currently doing is to try and reference the pronouns to their respective antecedents, for which I am trying to build a model. I have worked out the basic part…
user5794711
1
vote
0 answers

syntaxnet POS tagger models

I'm trying to make the (romanian) syntaxnet model run on a part of my corpus. I have the unzipped files mapped in a docker container. When i run the script for tokenizing: cat sent.txt | syntaxnet/models/parsey_universal/tokenize.sh…
omu_negru
  • 4,642
  • 4
  • 27
  • 38
1
vote
2 answers

Syntaxnet / Parsey McParseface declarative sentences only?

I cannot seem to find detail on the the data parsey was trained on; I am assuming it was mostly, if not all, on declarative sentences. If that's true, then Syntaxnet / parsey might not be the solution for imperative sentences, such as instructional…
Azeli
  • 698
  • 7
  • 16
1
vote
0 answers

How DocumentSink, DocumentSource ops in TF Syntaxnet actually work?

After going through the code... I could understand the flow as.. In parser_eval.py sink_documents = tf.placeholder(tf.string) sink = gen_parser_ops.document_sink(sink_documents, task_context=task_context, …
Aravind Pilla
  • 416
  • 4
  • 14
1
2