deep-text-recognition-benchmark icon indicating copy to clipboard operation
deep-text-recognition-benchmark copied to clipboard

model training

Open skruglov2023 opened this issue 2 years ago • 2 comments

I'm trying to train a model using my own images, for a specific purpose. I'm also new to deep learning algorithms and after reading multiple pages of conflicting training instructions for creating an easyocr model am more confused than I was previously. I've got a folder with plenty of images with text, and a text/csv file with every filename and the text found in the image.

How do I proceed from there? Is there a good video or text tutorial somewhere that can help? Or can someone help me train a model from scratch here?

skruglov2023 avatar Feb 17 '23 03:02 skruglov2023

create the dataset then train your model

https://github.com/clovaai/deep-text-recognition-benchmark/issues/85 read this

creati0nix avatar Feb 23 '23 11:02 creati0nix

I'm trying to train a model using my own images, for a specific purpose. I'm also new to deep learning algorithms and after reading multiple pages of conflicting training instructions for creating an easyocr model am more confused than I was previously. I've got a folder with plenty of images with text, and a text/csv file with every filename and the text found in the image.

How do I proceed from there? Is there a good video or text tutorial somewhere that can help? Or can someone help me train a model from scratch here?

can you please share the dataset that you have.

NeoDhirendra avatar Mar 21 '23 08:03 NeoDhirendra