train_classify.py

#!/usr/bin/env python3

import pandas as pd
import numpy as np
import warnings
import argparse
import matplotlib.pyplot as plt

from func.preprocess import *
from func.balance import *
from func.classifier import *
from func.feature_selection import *

pd.options.mode.chained_assignment = None
warnings.filterwarnings("ignore", category=DeprecationWarning)

parser = argparse.ArgumentParser()
parser.add_argument("-i", "--input", help="input file.",required=True)
parser.add_argument("-c", "--classifier",default='boot', help="classifier methods: kfold or boot")
parser.add_argument("-f", "--model", default='forest', help="feature selection \
                    model: logistic, forest, boosting")
parser.add_argument("-o", "--output", default='classify_model.sav', help="output model")
args = parser.parse_args()

# read in data
try:
    data = pd.read_csv(args.input, header=None)
    print("Data loaded")
except:
    print("Dataset could not be loaded. Is the dataset missing? Please use -i inputfile")

data.columns = ['AAGE', 'ACLSWKR', 'ADTIND', 'ADTOCC', 'AHGA', 'AHRSPAY',
                'AHSCOL', 'AMARITL', 'AMJIND', 'AMJOCC', 'ARACE', 'AREORGN',
                'ASEX', 'AUNMEM', 'AUNTYPE', 'AWKSTAT', 'CAPGAIN', 'CAPLOSS',
                'DIVVAL', 'FILESTAT', 'GRINREG', 'GRINST', 'HHDFMX', 'HHDREL',
                'MARSUPWT', 'MIGMTR1', 'MIGMTR3', 'MIGMTR4', 'MIGSAME', 'MIGSUN',
                'NOEMP', 'PARENT', 'PEFNTVTY', 'PEMNTVTY', 'PENATVTY', 'PRCITSHP',
                'SEOTR', 'VETQVA', 'VETYN', 'WKSWORK', 'YEAR', 'TARGET']

# preprocess data
data = preprocess(data)

# balance data by SMOTE
X, y, data = handle_imbalanced_data(data)

# drop instance weight from data
instance_weight = data.MARSUPWT
data = data.drop('MARSUPWT',axis=1)
# drop label from data
X = data.drop('TARGET',axis=1)



# methods of classifier
'''
method: Kfold_cross_validation, Bootstrap
'''
if args.classifier == 'kfold':
    warnings.warn("Kfold cross validation is only used to evaluate models, does not generate model file")

classifier_method(X,y,data,instance_weight,args.classifier,args.output,args.model)