Models

BERT models by Serhii Tiutiunnyk

multi_cased_bert_base_uk.zip — a multilingual cased base BERT model fine-tuned on a combined Ukrainian dataset (SQuAD 2.0 and SDSJ 2017) with the following parameters: epochs: 5, learning rate: 5e-5, max sequence length: 180, batch size: 16. Training used the script developed by the authors of the model (https://github.com/google-research/bert) and ran on an RTX 2080 Ti (11GB). On the squad-2.0-uk test set the model achieves 61.2% exact answers.

Permanent link

CLIP with Ukrainian support

CLIP model (ViT-H-14 + XLM-RoBERTa-large) supporting Ukrainian for zero-shot image classification.

Permanent link

Authors
Юрій Лаба, Володимир Мудрий
License
MIT
ELECTRA models for Ukrainian

A family of ELECTRA-base models pre-trained from scratch on Ukrainian texts (discriminators and a generator).

Permanent link

Authors
Stefan Schweter
License
MIT
Flair models for Ukrainian

Flair models for Ukrainian: named entity recognition (NER), POS tagging and contextual string embeddings.

Permanent link

Authors
Дмитро Чаплинський
License
MIT
NER models for MITIE

Thanks to our annotated corpus of Ukrainian we were able to train a model for automatic recognition of named entities (people, geographic names, company names, etc.) in unseen texts. For NER we chose the MITIE library. The library is open source, and its license permits free use even in commercial projects. MITIE also delivers high accuracy by combining familiar text features with CCA embeddings. Although MITIE is written in C++, it has bindings for other languages: C, Python, Java, Matlab. Please read the MITIE documentation and usage examples before you start.

To compute the CCA embeddings we used the corpus we collected of Ukrainian news, Wikipedia articles and fiction.

Download the NER model for Ukrainian

We also built a named entity recognition model for Russian, using the annotated corpus prepared by the organizers of the Dialogue 2016 conference. For its CCA embeddings we used a corpus of Russian Wikipedia articles.

Download the NER model for Russian

Permanent link

OmniGEC — grammatical error correction models

The OmniGEC family of grammatical error correction (GEC) models for Ukrainian.

Permanent link

Authors
Роман Ковальчук, Мар'яна Романишин, Петро Іванюк
License
MIT
Transformer-based spaCy NER models

Transformer-based spaCy NER models for Ukrainian (RoBERTa-large), trained on the NER-UK corpus.

Permanent link

Authors
Владислав Радченко, Назарій Друщак, Наталія Романишин, Дмитро Чаплинський
License
Apache-2.0 / MIT
Word embeddings (Word2Vec, GloVe, LexVec)

From the corpora we collected — news, articles, fiction, laws and legal texts — we computed the most widely used word embeddings: Word2Vec (and its improved version LexVec) and GloVe. We publish these models because computing them takes considerable time and server resources.

We built separate 300d models for every text category, and also computed them for lemmatized versions of the same corpora. If you need different model settings, take the corpora we prepared and compute the models to your needs.

With the help of Tetiana Kodliuk we created test sets, for evaluating the quality of the vectors, analogous to those that existed for English.
Quality evaluations were also performed; the results are available here.

word2vec parameters for small corpora (fiction):

./word2vec_standalone.py -size 300 -negative 7 -window 4 -threads 6 -min_count 10 -iter 5 -alpha 0.030

lexvec parameters for small corpora (fiction):

./lexvec -dim 300 -verbose 2 -negative 7 -subsample 1e-3 -window 4 -threads 6 -minfreq 10 -iterations 5 -alpha 0.030

GloVe parameters for small corpora (fiction):

MEMORY=4.0
VOCAB_MIN_COUNT=10
VECTOR_SIZE=300
MAX_ITER=15
WINDOW_SIZE=9
BINARY=2
NUM_THREADS=12
X_MAX=10:

For large corpora, minfreq is 25

Permanent link

fastText vectors for Ukrainian

fastText word vectors for Ukrainian, trained on the UberText 2.0 corpus.

Permanent link

Authors
Наталія Романишин, Дмитро Чаплинський, Кирило Захаров
License
MIT
Dragoman — English-Ukrainian translation model

State-of-the-art English-to-Ukrainian machine translation model (PEFT/LoRA on top of Mistral), with an online demo.

Permanent link

Authors
Юрій Панів, Дмитро Чаплинський, Нікіта Тринус, Володимир Кирилов
License
Apache-2.0
Tone dictionary expansion model

To expand the tone dictionary, Serhiy Shekhovtsov and Oles Petriv built a neural network model that finds words similar to existing ones using word2vec and lexvec vectors.

Examples of working with the model and its data:

Permanent link

Ukrainian sentence embeddings (sentence-transformers)

Sentence-transformers model for semantic embeddings of Ukrainian sentences, based on paraphrase-multilingual-mpnet.

Permanent link

Authors
Юрій Лаба, Володимир Мудрий, Дмитро Чаплинський, Мар'яна Романишин, Олесь Добосевич
License
Apache-2.0