How to use PaddlePaddle/PP-FormulaNet_plus-L with PaddleOCR:
# 1. See https://www.paddlepaddle.org.cn/en/install to install paddlepaddle # 2. pip install paddleocr from paddleocr import FormulaRecognition model = FormulaRecognition(model_name="PP-FormulaNet_plus-L") output = model.predict(input="path/to/image.png", batch_size=1) for res in output: res.print() res.save_to_img(save_path="./output/") res.save_to_json(save_path="./output/res.json")
