TypeError: object of type 'numpy.int32' has no len()

[belval]

TypeError: object of type 'numpy.int32' has no len()

[vinayakarannil]

i found a workaround. I commented sparse_tuple_from call from __generate_all_test_batches and used only batch_x and batch_y to run test. Anyways we just need batch_x only to do the prediction.

And yes, i had modified the filenames in the samples folder

Thanks

[Belval]

The images in the samples/ folder are the images displayed in the README.md, not valid testing examples as they do not have a ground truth in the filename.

[Belval]

You can use this dataset to test https://node1.belval.org/test_dataset.zip or generate a new one using https://github.com/Belval/TextRecognitionDataGenerator.