Split data into train/validation/test set

- [Instructor] Now that we have our clean data set, … this lesson should be quite simple. … We're just going to split up our full data set … so that we have 60% of our examples in the training set, … 20% in our validation set, … and 20% in the test set. … So import the packages that we'll need … and read in our data. … Again, we're using this train test split method … that we're importing from scikit learn. … And this is going to make our job very easy. … I also want to call out that we're reading in this … titanic cleaned data set that we created in the last lesson … rather than the original titanic.csv. … So go ahead and run that so. … Now, as we saw previously, we'll start by splitting … our data into our features … by just dropping the survived column. … So that will just leave the fields that are used to … make our prediction. … And then, we'll grab our labels here by just grabbing … this survived column. … I'm just going to highlight again before we jump in … that this is going to be a two-step process. …