I have a large set of images with 6 classes where the class distribution is imbalanced. Looked into balancing it by adding weight for each class but as I read in tensorflow docs the weight property is not taken into account i.e. any attribute beyond the set in use is simply ignored by TF2.

Since I am not using a custom written loop and would like to use training scripts present in TF2 models repository, I cant find any update info.

I already looked into the topic extensively on the web and found either outdated info or systems for integrating class weights into the training loop. Ideally I would like to add them either through pipeline.config or through the TF record formation process.

Can you help ?



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.