A neural parsing pipeline for segmentation, morphological tagging, dependency parsing and lemmatization with pre-trained models for more than 50 languages. Top ranker in the CoNLL-18 Shared Task.
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Parser-v2 @ 2af5e3b
docs
tokenizer @ 653e229
universal-lemmatizer @ eb4f9ea
.gitignore
.gitmodules
Dockerfile
Dockerfile.commonbase
LICENSE
README.md
build_lemma_cache.py
delexicalize_mod.py
dummy_handler.py
fetch_models.py
full_pipeline_server.py
full_pipeline_stream.py
lemma_cache_mod.py
lemmatizer_mod.py
marian_lemmatizer_mod.py
output_mod.py
parser_lib.py
parser_mod.py
pipeline.py
publish_model.sh
regextokenizer_mod.py
requirements-cpu.txt
requirements-gpu.txt
tokenizer_mod.py
tokenizer_udpipe_mod.py
wipe_mod.py
wstokenizer_mod.py

README.md

Turku-neural-parser-pipeline

A new take on the trusty old Finnish-dep-parser with pretrained models for more than 50 languages. The current pipeline is fully neural and has a substantially better accuracy in all layers of prediction (segmentation, morphological tagging, syntax, lemmatization).

Documentation: https://turkunlp.github.io/Turku-neural-parser-pipeline/