Training & Testing Data

Training and testing is an importan part of ML, do to so, we need some data to import, that the module has not seen. That is why we split the dataframe into training and testing parts.

If we would train our module on 100% of our dataframe, and than test it on the same - the results will always be perfect. But if we hide some data from him and show it later in the testing, we will get a more realistic picture.

Methods for Splitting Data

There are multiple ways to split our data into training and testing sets:

• Train-Test split

• Cross Validation