In [ ]:
'''Some backstory, I like to make automated and repeated purchases off a website and I like to login to a cashback portal first to get money back on my purchase. However, I see that the cashback website has now implemented recaptcha. So, it is time to see if I can crack it.'''
Start with basic approach from https://medium.com/@ageitgey/how-to-break-a-captcha-system-in-15-minutes-with-machine-learning-dbebb035a710
There's a google street view house number dataset. http://ufldl.stanford.edu/housenumbers/
There's a bunch of random characters dataset http://www.ee.surrey.ac.uk/CVSSP/demos/chars74k/
In [ ]: