pingmehard

Pinned Repositories

data-science-interviews
Data science interview questions and answers
Language:HTML8.7k 226 161.9k
bpe-dropout
Class allows you to create BPE tokens with dropout or not. Implements Sentencepeace lib with easy fit predict way.
Language:Python00
cherk
Language:Jupyter Notebook00
cian
Parser of cian.ru site and modeling classification to devide photos of the resource into telegramm channel.
Language:Jupyter Notebook00
data-science-interviews
Data science interview questions and answers
Language:HTML00
newsbot
News Bot allows you to choose and personalize news in one flow.
Language:Python00
stratify
Code allows you to strtify dataframe in different convenient ways. ReadMe shows how.
Language:Jupyter Notebook00
tfidf
TF-IDF is a method, which allows get matrix tokens for your list of text. It creates n * m matrix, where n - quantity of texts and m - quantity of unique words - tokens in texts vocabulary. IMPORTANT: This .py is my own variation of tfidf and doesnt duplicate existing tfidf versions. So it could
Language:Python00
tinkoff-api
tinkoff-api
Language:Python00
word2vec
Word2Vec is an algorythm of word representation in embeddings. This repo contains a code about word2vec only.
Language:Python00

pingmehard's Repositories

pingmehard/bpe-dropout
Class allows you to create BPE tokens with dropout or not. Implements Sentencepeace lib with easy fit predict way.
Language:Python00
pingmehard/cherk
Language:Jupyter Notebook00
pingmehard/cian
Parser of cian.ru site and modeling classification to devide photos of the resource into telegramm channel.
Language:Jupyter Notebook00
pingmehard/data-science-interviews
Data science interview questions and answers
Language:HTML00
pingmehard/newsbot
News Bot allows you to choose and personalize news in one flow.
Language:Python00
pingmehard/stratify
Code allows you to strtify dataframe in different convenient ways. ReadMe shows how.
Language:Jupyter Notebook00
pingmehard/tfidf
TF-IDF is a method, which allows get matrix tokens for your list of text. It creates n * m matrix, where n - quantity of texts and m - quantity of unique words - tokens in texts vocabulary. IMPORTANT: This .py is my own variation of tfidf and doesnt duplicate existing tfidf versions. So it could
Language:Python00
pingmehard/tinkoff-api
tinkoff-api
Language:Python00
pingmehard/word2vec
Word2Vec is an algorythm of word representation in embeddings. This repo contains a code about word2vec only.
Language:Python00