Tatiana Dembelova

Published in Towards Data Science

·Dec 6, 2020

Part 1. OCR Typo detection with BERT embeddings, CNN and PyTorch

OCR Typo correction series — Introduction In this series I would like to present a solution for the OCR typo correction task. This problem is not new: unfortunately, many of us encountered bad quality of the OCRed text, especially if the scanned document had unusual font, some scan artefacts or contained unknown words. It is possible…

Ocr

7 min read

Ocr

7 min read

Published in Towards Data Science

·Aug 20, 2020

Monty Hall problem solution using Bayes theorem

Intro A couple of days ago I decided to refresh my knowledge of probability theory, so the book “All of Statistics” from Larry Wassermann came quite handy at that point. Somewhere in the first chapters of the book the catchy Monty Hall problem got my attention. Actually, I knew already about…

Bayes Theorem

7 min read

Bayes Theorem

7 min read

Published in Towards Data Science

·Aug 6, 2020

Introduction to generative and discriminative models

Intro Recently I gave a presentation at work, where I explained how I solved some problem using Conditional Random Fields (CRF). And, since CRF was not so much known to my colleagues, I had a couple of slides devoted to the theory behind this algorithm. As I prepared the theory…

Generative Model

4 min read

Introduction to generative and discriminative models

Generative Model

4 min read