search

awslabs / handwritten-text-recognition-for-apache-mxnet

This repository lets you train neural networks models for performing end-to-end full-page handwriting recognition using the Apache MXNet deep learning frameworks on the IAM Dataset.
Apache License 2.0
481 stars 189 forks source link

How to make predictions from pre-trained models? #36

Closed naveen-marthala closed 4 years ago

naveen-marthala commented 4 years ago

Good work, thanks. I am using pre-trained models to get text from images. When I was going through the codes on how to do it, I learned that my test images's format has to match the format 'IAMDastaset' class in 'ocr.utils.iam_dataset' outputs. So, How do I modify 'IAMDastaset' class in 'ocr.utils.iam_dataset' to change an input image to match test dataframe format that this class outputs or How do I get dataframe for images other than the ones in IAM dataset. I couldn't understand this class completely. So, if anyone worked on this, please help me solve this.

jonomon commented 4 years ago

Hi Naveen-kumar-123,

You can follow this tutorial to create your own dataset.

jbuehler1337 commented 4 years ago

Hey, how can I just upload a single image to make a prediction on it? It would be crazy to make a whole dataset for a single prediction.

jonomon commented 4 years ago

You can edit the images array in https://github.com/awslabs/handwritten-text-recognition-for-apache-mxnet/blob/master/0_handwriting_ocr.ipynb

naveen-marthala commented 4 years ago

@jonomon, I will try that and get back to you very soon. Please do not close this issue before that.

naveen-marthala commented 4 years ago

@jonomon As you had suggested, I have created images array like: images = [cv2.imread('/location of image/image_1.jpg'), cv2.imread('/location of image/image_2.jpg')] This worked. Thanks

jonomon commented 4 years ago

You're welcome