I have used the feature-processor (with preprocess_test_data) to zscore-normalize the test data based on the mean and std of the training data.
I would like the classifier to use these zscored-normalized data but I am not sure of how do that. I have indeed thought of redefining a ds object but, from my understanding, the test and training data will then be put altogether, while I would have liked to keep the test/training data… I thought of giving the feature-processor as an input to the classifier as well but it seems to me that it contains information relative to the mean and std of the training data, not of the test data.
So, I do not understand how I can use my preprocessed (ie zscored normalized) test data.
Thank you in advance and sorry if this question is naive – I am a total newbie.
This topic was modified 8 years, 4 months ago by dartvhunger.