Simplify Latex OCR

Quick Start Training and Evaluation

This repo is an open resource image-to-latex project based on pix2tex.

Install

pip install -r requirements.txt

Download Model

Image Resizer Model

wget https://drive.google.com/file/d/1U1raFeRG2LUCSkF0ozu2ByOOmu3lAJHF/view?usp=drive_link

Latex OCR Model

wget https://drive.google.com/file/d/1dTVdfl7CU8YQxRxhBwfm8lwgH-IQovxi/view?usp=drive_link

Run Code

See playground.ipynb

Training

from main.utils.jsonData import JsonData
from main.trainer import Trainer

config = JsonData('./config/colab.json')

trainer = Trainer(model_config=config, vocab_file='./model/tokenizer.json', train_data_path=[...], eval_data_path=[...], resume_path='...', batchsize=30, pad=True, task_name='...')

for i in trainer(num_epochs=120, lr=1e-4):
    a = i

Pred

from main.utils.jsonData import JsonData
from main.predictor import LatexOCR

config = JsonData('./config/colab.json')

model = LatexOCR(model_config=config, vocab_file='./model/tokenizer.json', resume_path='./save_model/released/Checkpoint_8146/model.pth', resize_model_path='./model/image_resizer.pth')

import os
from PIL import Image

dir = './test_samples'
imgs = ['1.png', 'a.png', 'b.png', 'c.png', 'd.png', 'e.png', 'f.png', 'g.png', 'h.png', 'i.png', 'j.png']

for path in imgs:
    img = Image.open(os.path.join(dir, path))
    math = model(img)
    print(f'### {path}\n${math}$\n')

API

Make Config

cp config/app_config.json.example api/app_config.json

You can set you own api_key in api/app_config.json, otherwise, the app will run without authentication.

Start Server

cd api/

uvicorn app:app --host=0.0.0.0 --port=8000

Docker

Build and run the docker image

docker compose up -d --build

The code is deploy based on shared storage, so that you can modify the code in the physical machine and the docker container will automatically update the code. So, do not delete any necessary code files in the physical machine.

Restart the docker container

docker compose restart

License

MIT License

Copyright (c) 2024 Creator SN®

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.