Closed miriam-agrawala closed 9 months ago
I think because our objective at the end is to normalize the feature values before feeding it into the model anyway, it wouldn't make that much of a difference. But in terms of which is better I agree with Adham in that floats can be converted to integers but not the other way round. So if there is a choice, lets preferably go with floats.
we discussed this in the project group and decided to go with floats
I think float is more accurate than ints. But if all values are only in one decimal point (eg: 1.0, 3.0, 14.0, 20.0), then we can convert it to ints. For now, I will stick to float first