Question about retrain your model on PDBbind-2016 by modifying the 'act' column

[RJ-Li]

Hi, I am new to machine-learning scoring functions, and I want to use RFScore to re-train your model on PDBbind_v2016.
I have read the csv of v1 descriptors:

pdbid,2016_general,2016_refined,2016_core,2015_general,2015_refined,2015_core,2014_general,2014_refined,2014_core,2013_general,2013_refined,2013_core,2012_general,2012_refined,2012_core,2007_general,2007_refined,2007_core,act,0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35
10gs,True,True,False,True,True,True,True,True,True,True,True,True,True,True,False,True,True,True,6.4,2877,717,1030,37,408,105,160,7,747,180,288,12,0,0,0,0,0,0,0,0,146,39,56,3,0,0,0,0,0,0,0,0,0,0,0,0
11gs,True,False,False,True,False,False,True,False,False,True,False,False,True,True,False,True,True,False,5.82,3137,803,1088,41,413,110,154,8,1191,286,425,14,0,0,0,0,0,0,0,0,165,39,59,3,286,71,85,2,0,0,0,0,0,0,0,0
184l,True,True,False,True,True,False,True,True,False,False,False,False,False,False,False,False,False,False,4.72,1939,455,483,24,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0

Now I want to ask if I could modify this csv's 'act' column to train a new model.pkl based on PDBbind-2016? How could I rename the modified csv and set it in the script?
The following is what I learned when training RFScore model:

from RFScore import rfscore
model = rfscore(protein=your_protein, n_jobs=-1, version=1, spr=0)
model.gen_training_data(pdbbind_dir=your_dir, pdbbind_versions=(2007, 2012, 2013, 2014, 2015, 2016))
model.train(home_dir=your_home_dir, pdbbind_version=2016)