I am working with GNNs for a node classification problem, and I am only able to achieve about 50% accuracy for the training set. I am not able to overfit the model to the training data. This makes me think that the feature space does not sufficient describe the data I am trying to classify.

Is there a test to check if the feature space is sufficient for a problem? I expected overfitting to the training data to be possible, but I am wondering if my feature are just not adequate.



You must log in to answer this question.