transformer-pytorch VS LaTeX-OCR

In the original paper they mention they are only using the decoder part of the model. However, their description and implementations seem to be using the encoder part of the transformer not the encoder. For example, this implementation of the original transformer encoder layer matches what the one in the GPT implementation.

I implemented the Transformer model of Google Brain using Pytorch. It was specially written together in very detailed and easy explanatory comments. If you're a beginner who wants to implement Transformer, look at my code and try it! Detailed code can be found here. (https://github.com/hyunwoongko/transformer-pytorch)

I didn't know that such tools exists (completly new to LaTex). Thanks to your suggestion I looked for an open source althernative (to avoid anoyances of freemium) and I found pix2tex That works really like a charm.

I have recently discovered a few very helpful github packages which help me make notes while listening to lectures. These would be 1. pix2tex (allows you to scan an equation and convert it to latex) 2. pix2text (allows you to scan an equation with words in it and converts it to latex and text) 3. Tesseract (not really a physics related package, but it does allow me to copy notes from transcripts easily) 4. Mathpix an app that performs all the above mentioned operations better than the packages above, but one which ain't free.

This is also a great aid to learing LaTex. I wonder if anyone has ever tried to make an OCR system that generates the appropriate LaTex from an picture of an equation?

Turns out the answer is yes:

https://github.com/lukas-blecher/LaTeX-OCR

LaTeX-OCR