Data Preparation

`

1. Sample and Convert tif to png and resize - sanskrit-handwritten-ocr/training_data_prep/images_sampling.py
2. Dump pcikle files from Google-Cloud_vision - sanskrit-handwritten-ocr/training_data_prep/google_ocr_response_dump.py
3. Create Preannotations CVAT format - sanskrit-handwritten-ocr/training_data_prep/create_preannotations_cvat.py
4. Upload annotations to CVAT and correct the annotations
5. Split the data and annotations - sanskrit-handwritten-ocr/training_data_prep/ppocr_data_prep.py
6. Reformat to PPOCR or EasyOCR- sanskrit-handwritten-ocr/training_data_prep/ppocr_data_prep.py