i’m new to ml and i’m trying my hand at mlr to predict a student’s final grade using this kaggle dataset

i know that you’re supposed to
a) encode all binary and categorical data
b) discard variables displaying multicollinearity
c) pick variables that have a linear relationship with your dependent (i’m a bit confused about this as well since in a lot of videos i’ve watched so far, people don’t really check for this when training their model using linear regression)

i have done (a) by labelencoding my binary variables, and onehotencoding my categorical variables. i even calculated my vif for each (the onehotencoded ones have an infinite vif value). and now… i’m not sure how to proceed. all i can safely say is that since G2 has the high vif score, i can discard it; and that since Medu has a close but higher score than Fedu, i can discard that too (same with Walc and Dalc)

here are the value i’m getting:

const                 0.000000
school                1.518331
sex                   1.489316
age                   1.818399
address               1.388570
famsize               1.153361
Pstatus               1.145962
Medu                  2.946452
Fedu                  2.147572
traveltime            1.322387
studytime             1.398220
failures              1.567588
schoolsup             1.262329
famsup                1.306325
paid                  1.339139
activities            1.167950
nursery               1.153852
higher                1.316551
internet              1.258651
romantic              1.179480
famrel                1.173444
freetime              1.322079
goout                 1.496537
Dalc                  2.036903
Walc                  2.405555
health                1.181635
absences              1.297898
G1                    4.794857
G2                    8.414788
G3                    6.483623
Mjob__at_home              inf
Mjob__health               inf
Mjob__other                inf
Mjob__services             inf
Mjob__teacher              inf
Fjob__at_home              inf
Fjob__health               inf
Fjob__other                inf
Fjob__services             inf
Fjob__teacher              inf
reason__course             inf
reason__home               inf
reason__other              inf
reason__reputation         inf
guardian__father           inf
guardian__mother           inf
guardian__other            inf
dtype: float64