Question: Image classification neural network performs well on python generated data but poorly on real data

Follow the full discussion on Reddit.
Hi. Im a student trying to make a neural network that can classify light diffraction images created with python. The problem is that the model achieves very high scores on accurasy and the classification report/confusion matrix when tested on new pythong generated images but it performs very poorly when i give it real life images i find from the internet. I suspect it has to do with the less than ideal noise the real images have and now im trying to introduce noise to my training dataset. Any tips? Bellow is an single slit diffraction image generated with python (images like that are on the training set) real images of diffraction will have different colors and more noise in the background. I try to augment my train dataset but i dont know if it will work.

Comments

There's unfortunately not much to read here yet...

Discover the Best of Machine Learning.

Ever having issues keeping up with everything that's going on in Machine Learning? That's where we help. We're sending out a weekly digest, highlighting the Best of Machine Learning.

Join over 900 Machine Learning Engineers receiving our weekly digest.

Best of Machine LearningBest of Machine Learning

Discover the best guides, books, papers and news in Machine Learning, once per week.

Twitter