In machine learning training, we’re looking for features with a balanced distributions of values. Basically, you want the feature histogram to look like a symmetric hump, with flanks on the left and right tapering to zero.
But do you remember the histogram of the median_house_value column from the previous lab lesson? It looks like this:
...
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