I have the training data in the database (mysql), is there any best practice as to, using the training/testing data as a CSV vs reading from database query ?
You must be logged in to post. Please login or register an account.
Whether I had a CSV or in a database, I would load both into a Pandas dataframe anyway, so it makes no difference. At some point, the data is for sure becoming a Numpy array, so, again, makes no difference. I like to load from original source into Pandas first so I can do organization and some manipulation before converting to a numpy array, but you may not need that step.
-Harrison 8 years ago
You must be logged in to post. Please login or register an account.