Comments (3)
You cannot, unfortunately, select individual components of a stored pipeline easily. However, you could adjust the config.json file of the corresponding model to include only the connective component (just remove the other entries that follow).
For data preparation, I recommend using discopy-data in combination with the parser as follows (I guess the interface changed over the last days):
- traditional feature-based parser:
discopy-tokenize -i examples/wsj_0336 | discopy-add-parses -c | discopy-parse lin models/lin
- neural parser:
discopy-tokenize --tokenize-only -i examples/wsj_0336 | discopy-nn-parse bert-base-cased models/pipeline-bert-2
You could also first process your documents, store them in a single file and then parse this file by either parser pipeline:
discopy-tokenize -i examples/wsj_0336 | discopy-add-parses -c > dataset.json
discopy-tokenize -i examples/wsj_034X | discopy-add-parses -c >> dataset.json
discopy-nn-parse bert-base-cased models/pipeline-bert-2 -i dataset.json
Note: discopy-tokenize
splits text files into multiple documents by three or more consecutive newlines (r'\n\n\n+')
from discopy.
You cannot, unfortunately, select individual components of a stored pipeline easily. However, you could adjust the config.json file of the corresponding model to include only the connective component (just remove the other entries that follow).
For data preparation, I recommend using discopy-data in combination with the parser as follows (I guess the interface changed over the last days):
- traditional feature-based parser:
discopy-tokenize -i examples/wsj_0336 | discopy-add-parses -c | discopy-parse lin models/lin
- neural parser:
discopy-tokenize --tokenize-only -i examples/wsj_0336 | discopy-nn-parse bert-base-cased models/pipeline-bert-2
You could also first process your documents, store them in a single file and then parse this file by either parser pipeline:
discopy-tokenize -i examples/wsj_0336 | discopy-add-parses -c > dataset.json
discopy-tokenize -i examples/wsj_034X | discopy-add-parses -c >> dataset.json
discopy-nn-parse bert-base-cased models/pipeline-bert-2 -i dataset.json
Note:
discopy-tokenize
splits text files into multiple documents by three or more consecutive newlines (r'\n\n\n+')
Thanks for your response. I have been able to run first two steps successfully. However, this repository doesn't contain 'models' folder, as well as config files. Can you please upload those too?
from discopy.
To not blow up the repository with models' histories, I put them under releases:
https://github.com/rknaebel/discopy/releases
from discopy.
Related Issues (19)
- Add support for general text files
- unsatisfying results on argument extraction
- Format of input data HOT 1
- Run on GPU
- prediction doesn't work HOT 1
- CRF based Argument extraction
- train data json explanation HOT 1
- Installation problem HOT 2
- training data HOT 2
- Convert PDTB data into conll
- Convert PDTB2 Data to JSON Format for Discourse Parser Training HOT 1
- Extend implicit relation classification HOT 1
- Add confidence score for relation prediction
- Helpful README with usage examples
- Add different levels of sense classification
- Add Dimlex support for Connective Classifier
- add CoNLL exact evaluation functions
- alternative for connective head mapper
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from discopy.