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

OSError #52

Closed JPremnath06 closed 3 years ago

JPremnath06 commented 3 years ago

getting the below error while executing dataset creation step. train_ds = IAMDataset("form", output_data="bb", output_parse_method="form", train=True)

jonomon commented 3 years ago

Hi @JPremnath06 What is the error message?

JPremnath06 commented 3 years ago

image Attached the error i got

JPremnath06 commented 3 years ago

As per my understanding..instead of datset the web page is downloading as .tgz file. I have tried with both credentials.json file and credentials as parameter not working. Please suggest

jonomon commented 3 years ago

Could you try to open xml.tgz in your OS to see if it was downloaded correctly?

JPremnath06 commented 3 years ago

Yeah I tried that and it shows as a HTML page aroud 4.1 kb. The dataset is not download using the link provided. I have tried it externally for the login page it shows internal server error occurred or 404 not found

jonomon commented 3 years ago

We do not have access to the IAM dataset. You may have to contact the dataset owner to resolve your issues with downloading the dataset.

harshsp31 commented 3 years ago

Hi @jonomon, I have the same issue and I identified the cause, the URLs on which the dataset has been hosted have been changed. I adjusted for the URLs for most cases but I'm unable to find a URL for "http://www.fki.inf.unibe.ch/DBs/iamDB/tasks/largeWriterIndependentTextLineRecognitionTask.zip". How do I solve this issue?

jonomon commented 3 years ago

Hi @harshsp31,

Thank you so much. Would it be possible for your to submit a PR with the updated URLs?

harshsp31 commented 3 years ago

Sure I will do that!

janijanmejay commented 3 years ago

@harshsp31 the new URL that you are missing for largeWriterIndependentTextLineRecognitionTask.zip is https://fki.tic.heia-fr.ch/static/zip/largeWriterIndependentTextLineRecognitionTask.zip

RohanVKashyap commented 3 years ago

@harshsp31 Could you raise the PR regarding the changes in the url 's to be made,I have also encountered the same error.