Hello, I’m a medical student working on a radiology research project where I’m trying to train a CNN model to identify and stage diseases from X-rays.
When I execute the model, it predicts only one Class.
Any advice on how to cope with this bug, and how should xrays be preprocessed in Knime?
It’s a little hard to guess based on just screenshots, but are you only predicting classification for a dataset of size n=4? In general, what are the sizes of your training, validation, and test sets?
If your data is not confidential, uploading a sample of it along with your workflow to the KNIME Community Hub would help others here to better diagnose your problem.