Discover ufal/weighteddist Open Source project by ÚFAL (@ufal)

An open-source tool for sequence learning in NLP built on TensorFlow.

1,782

neuralmonkey

UDPipe: Trainable pipeline for tokenizing, tagging, lemmatizing and parsing Universal Treebanks and other CoNLL-U files

410

udpipe

Source code for paper Neural Architectures for Nested NER through Linearization

344

acl2019_nested_ner

Embeddable C++17 Unicode library offering UTF encodings, general category info, simple and full casing, normalization forms, and combining marks stripping.

unilib

MorphoDiTa: Morphologic Dictionary and Tagger

morphodita

Tool that will help you select the right open license for your data or software

public-license-selector

CoffeeScript

perin

PERIN is Permutation-Invariant Semantic Parser developed for MRP 2020

NameTag: Named Entity Tagger

nametag

Distributed infrastructure for Machine Translation web services (using Moses, Python, JSON-RPC/web interface)

mtmonkey

treex

Treex NLP framework

Materials for the Deep Learning -- ÚFAL course NPFL114

npfl114

npfl129

NPFL129 repository

Frontend of LINDAT translation service

lindat-translation

Lightweight self-hosted span annotation tool

factgenie

augpt

DSTC9 Submission

Statistical spell- and (occasional) grammar-checker.

korektor

Deep Learning Seminar -- ÚFAL course NPFL117

npfl117

multilexnorm2021

MultiLexNorm 2021 competition system from ÚFAL

Parsito: Fast non-projective transition-based dependency parser

parsito

npfl122

NPFL122 repository

MicroRestD is a small C++11 cross-platform REST server built on top of libmicrohttpd http://www.gnu.org/software/libmicrohttpd/.

microrestd

Source code for paper Grammatical Error Correction in Low-Resource Scenarios (W-NUT 2019)

low-resource-gec-wnut2019

A system for live lecture translation (speech to text) where the audience can easily provide corrections.

correctable-lecture-translator

Practical End-to-End Optical Music Recognition for Pianoform Music

olimpic-icdar24

A minimal Python implementation of the Treex API

pytreex

LinPipe: Multilingual Processing Tool

linpipe

nlgi_eval

NLI evaluation for NLG

Chu-Liu-Edmonds maximum spanning algorithm from TurboParser for use within Python

chu_liu_edmonds

a simple tool to parse marian training logs and display them in tensorboard

marian-tensorboard

UFAL-Prague entry to the Sigmorphon 2019 Shared Task 2

sigmorphon2019

Makefiles, scenarios and support scripts for the development of HamleDT within the Treex infrastructure

hamledt

Makefile

wnut2021_character_transformations_gec

The code from the paper Character Transformations for Non-Autoregressive GEC Tagging

LINDAT/CLARIN repository for linguistics (http://lindat.cz)

lindat-repository-obsolete

Java

charles-translator-web-frontend

Charles Translator: MT from Charles University

TypeScript

clarin-sp-aaggregator

PHP

mrpipe-conll2019

ÚFAL MRPipe submission to CoNLL 2019 shared task

SliMD presentation system based on Markdown and HTML5&js.

slimd

Build scripts for the UniSegments collection of morphologically segmented lexicons for many languages

universal-segmentations

Latex repository for a poster design

UFAL_poster

TeX

bert-diacritics-restoration

Repository storing code and data for our paper "Diacritics Restoration using BERT with Analysis on Czech language".

EACL 2023 paper "MLASK: Multimodal Summarization of Video-based News Articles"

MLASK

LatinPipe – the winning entry to parsing task of EvaLatin 2024

evalatin2024-latinpipe

optimal-reference-translations

CoNLL 2017 Shared Task Proposal: UD End-to-End parsing

conll2017

Scripts for extracting errors from Wikipedia revisions

wiki-error-corpus

ÚFAL Thesis Information Repository

ÚFAL Reading Group

thesis_info

Query engine and query language for trees in PML format

perl-pmltq

Reading Hackathon -- NN Tagging Project

rh_nntagging

Refactored and simplified PMLTQ::CGI

perl-pmltq-server

Coreference extension to Prague Czech-English Dependency Treebank 2.0

pcedt2.0-coref

Makefile

kazitext

Coreference and anaphora scorer for CorefUD data

corefud-scorer

A handy tool for quick manual evaluation of line-oriented outputs, e.g. of machine translation.

quickjudge

Conversion tools to and from the TEITOK TEI/XML format

teitok-tools

CoNLL 2018 UD Shared Task

conll2018

Android app of LINDAT translation service

charles-translator-android

Kotlin

crac2023-corpipe

ÚFAL CorPipe: CRAC 2023 Winning System for Multilingual Coreference Resolution

QTLeap Pilot MT systems using TectoMT

qtleap

Consolidated Czech PDT-style annotated corpus; consists of PDT, Czech part of PCEDT, PDTSC, PDT-Faust

PDT-C

lindat-corpora-conversions

LINDAT Corpora Conversions

Parse shibboleth logs for important information about attributes from IdPs and other

lindat-aai-attributes

XSLT

ufal-tools

Delexicalized tagging and parsing.

deltacorpus

Javascript library for visualizing Treex files

js-treex-view

A template PhD thesis at UFAL

phd-thesis-template

TeX

cpp_builtem

C++ Builtem is a cross-platform Makefile-based build system for C++11

Shell

ambiguity-grammaticality-complexity

Code for the paper Sentence Ambiguity, Grammaticality and Complexity Probes

Common files and branding for Lindat projects

lindat-common

ÚFAL CorPipe: CRAC 2022 Winning System for Multilingual Coreference Resolution

crac2022-corpipe

Cashing important counts from PIWIK periodically and creating customized reports for LINDAT/CLARIN

lindat_piwik_reports

EMMT (Eyetracked Multi-Modal Translation), a simultaneous eye-tracking, 4-electrode EEG and audio corpus for multi-modal reading and translation scenarios

eyetracked-multi-modal-translation

uk-cs-data-scripts

Scripts for processing data for Czech-Ukrainian MT

errant_czech

UFAL_MT_service

NameTag3: Named Entity Tagger

nametag3

mrptask

Python package using clarin-dspace API

lindat-aai-discovery

HTML

pyclarindspace

Jupyter Notebook

ParCzech

ParCzech is a project on compiling Czech parliamentary data into annotated corpora.

GLSL

theaitrobot

THEaiTRE bot

Semantic MT metric trained on HUME annotations

auto-hume

Repository of the seminar NPFL101 Competing in Machine Translation.

npfl101

Shell

bilingual-abstracts-corpus

Bilingual corpus of scientific abstracts from ÚFAL Charles University publications.

Experiment relating NMT quality and post-editing efforts

continuous-rating

PHP

tamiltb

Makefile

nmt-pe-effects-2021

MTEQA

cpp_utils

europarlmin

Corpus of European Parliament debates organized as a corpus for meeting summarization, i.e. matching full transcripts and minutes from the sessions. Used in the shared task of AutoMin 2023.

pmltq-cgi

PMLTQ::CGI has been removed from PMLTQ module in order to decrease number of dependencies. It should be installed separately.

A wrapper over various grid submission scripts

qsubmit

SynSemClassSearch

Simple Czech and English keyword extractor

ker

NPFL087 Statistical Machine Translation

npfl087

Shell

diaser

Online interface for Treex

treex-web

TF2 service for word embeddings computation

wembedding_service