search

Mushroomcat9998 / PaddleOCR

Custom repo for training Japanese OCR
Apache License 2.0
18 stars 10 forks source link

Paddle OCR Text detection Model Evaluation take too much time #3

Open kokul93 opened 2 months ago

kokul93 commented 2 months ago

I am training PaddleOCR for Tamil language and when I train the model with 70 images for training and 24 images for evaluation.

After 1st epoch the model take more time (More than 5 hrs) on Evaluation

Mushroomcat9998 commented 2 weeks ago

I have no idea why it takes a long time for Text detection Model Evaluation with PaddleOCR, but I think your dataset when training has some problems. I have often used DBNet repo for training Text Detection Model, you can try it if you find it interesting. Thanks for watching my custom repo