imvladikon Goto Github PK

followers: 56.0 following: 709.0 repos: 52.0 gists: 32.0

Name: Vladimir Gurevich

Type: User

Twitter: VladimirGurevi2

Location: Israel

Hi

I'm Vladimir Gurevich, ML/NLP Engineer (IR tasks, such as Semantic Search, Information Extraction tasks, such as NER, Relation Extraction, etc.).

I am also interested in Speech Recognition and in LLMs.

Works:

jupyter-notebook-viewer - Jupyter Notebook Viewer for local files *.ipynb in browser without Jupyter Notebook installation.
wav2vec2-hebrew - package for speech recognition in Hebrew language using wav2vec2 models that were trained on Hebrew datasets (check out the datasets below).
distiller - distillation TextClassification and TokenClassification models using transformers library with different distillation methods.
spacy-trankit - spacy wrapper for Trankit (NLP pipeline for tokenization+dependency parsing+lemmatization, etc.)

Models:

t5-english-ner - NER model that based on T5 encoder that was trained on extremely small dataset.
sentence-transformers-alephbert - Sentence Transformers model that based on AlephBERT model for sentence similarity tasks.
het5_small_summarization - mt5-small based summarization model for Hebrew

Speech Recognition:

Datasets:

Contacts

Vladimir Gurevich's Projects

abydos

Abydos NLP/IR library for Python [imvladikon] made some changes

ankitools4j

anki decks creation in Java

annotations_deduplications

scripts to deduplicate annotations and to refine NER spans or to analyze the differences

babylon-parser

convert babylon (dgl) dictionaries to html in Java

biorxiv_scraper

bm25_vectorizer

sklearn compatible bm25 vectorizers

campus-dl

A simple tool to download video lectures from campus.gov.il (based on edx-dl)

cdatasketch

MinHash, LSH, LSH Forest, Weighted MinHash, HyperLogLog, HyperLogLog++, LSH Ensemble [imvladikon] added cython implementations

character-bert

Main repository for "CharacterBERT: Reconciling ELMo and BERT for Word-Level Open-Vocabulary Representations From Characters"

character-bert-pretraining

Code for pre-training CharacterBERT models (as well as BERT models).

charbert

CharBERT: Character-aware Pre-trained Language Model (COLING2020)

datasets

🤗 Fast, efficient, open-access datasets and evaluation metrics in PyTorch, TensorFlow, NumPy and Pandas

datasets_additional

deduplicator

Simple entity deduplication package

deep_learning_school

tasks and projects from the deep learning school by MIPT

distil-whisper

Distilled variant of Whisper for speech recognition. 6x faster, 50% smaller, within 1% word error rate.

distiller

knowledge distillations for bert (classification, token classification models)

easyocr

Ready-to-use OCR with 40+ languages supported including Chinese, Japanese, Korean and Thai

evaluate

🤗 Evaluate: A library for easily evaluating machine learning models and datasets.

events_leaflet

playing with leaflet, events

hebrew_summarizer

finetuning experiments on summarization tasks for Hebrew

html_extractor

huawei-nlpcourse-project

Topic modeling and classification news on Hebrew with Neural Text Summarizer model

imvladikon

indonesian_nlp_experiments

some experiments in Indonesian NLP (information extraction from the courts reports)

jupyter-notebook-viewer

chrome extension for viewing Jupyter Notebooks in the browser without Jupyter Server

kdbush

Naive implementation of Java KD-Bush(KD-tree) algorithm

knesset-2004-2005

A corpus of transcriptions of Knesset (Israeli parliament) meetings between January 2004 and November 2005.

nbpreview

Render IPython/Jupyter notebooks without running a notebook server.

news_scrapers

This repository contains scripts for scraping news from different sources