I'm using the AutoML API for ML.Net, and I see from the docs that it does a bunch of data pre-processing for me. What I don't see though is if it normalizes each of the feature columns for me or not. My question is: Does AutoML normalize my data for me? If so, do I have any control as to how it normalizes? If not, how would I do this given that I'm not creating a pipeline that I see all the other normalization examples use?
From my experience AutoML engine automatically adds affine normalizer to Ols regression models, all other types don't get any normalization. You can add normalizer as the preFeaturizer argument to the Execute method, e.g.
preFeaturizer: mlContext.Transforms.NormalizeLpNorm(featuresColumnName, featuresColumnName)