[Python] Need to keep data in two columns connected, while repeating across 100 rows

I am working on a data set to prepare it for modeling in python. Here is my issue: I have a group of two columns that I need to keep connected (ex. a = ...) but repeat it throughout the data set. See below:

col 1 col 2 col 3
-0.363965 a 1.04344
0.566 b -1.042184
-1.734601 c -1.072206
-0.944462 d -0.981626
-1.15268 e -1.060917
-1.734601
-0.944462
-1.15268
-0.489548
0.682
0.698
0.404
0.852
0.450439
0.630066
0.564755
0.534752

I've tried pd.merge, np.repeat, etc. All I can get is either repetition of the first value in column 3 or NaN or errors. Is this even possible in python?

Ideally, it would look like this:

col 1 col 2 col 3
-0.363965 a 1.04344
0.566 b -1.042184
-1.734601 c -1.072206
-0.944462 d -0.981626
-1.15268 e -1.060917
-1.734601 b -1.042184
-0.944462 b -1.042184
-1.15268 c -1.072206
-0.489548 a 1.04344
0.682 c -1.072206
0.698 e -1.060917
0.404 a 1.04344
0.852 b -1.042184
0.450439 c -1.072206
0.630066 a 1.04344
0.564755 e -1.060917
0.534752 a 1.04344

I've tried pd.merge, np.repeat, etc. All I can get is either repetition of the first value in column 3 or NaN or errors. Is this even possible in python?

For additional context - column 3 are the dummy variables created for the category of column 2. I want to be be able to model column 1 values (that are associated with column 2 categories) in a linear regression model with other interactions and additives (all continuous) and show the predicted performance with confidence interval and prediction intervals in a graph.

Continue reading...