from dataPreProcessing import process
from featureEngineering import featureProcessing
from featureSelection import featureSelecting
from dataTransformation import dataTransform
from modelTraining import model
#del sys.modules['logBuilder']
#del sys.modules['dataIngestion']
#del sys.modules['dataPreProcessing']
#del sys.modules['featureEngineering']
#del sys.modules['featureSelection']
#del sys.modules['dataTransformation']
#del sys.modules['modelTraining']
fileDir = os.path.dirname(__file__)
sys.path.insert(0, fileDir)
updatelog = AppLogger()
try:
import pandas as pd
from sklearn.model_selection import train_test_split
import time
updatelog.log('process', 'Imported necessary packages')
except:
updatelog.log(
'error',
"Unexpected error:" + str(sys.exc_info()[0]) + str(sys.exc_info()[1]))
masterBegin = time.time()
# Database ingestion
try:
dataLoad = loadData.Load()
"""
Created on Fri Aug 21 20:02:24 2020
@author: raghu
"""
import sys
import os
import numpy as np
fileDir = os.path.dirname(__file__)
dirPath = os.path.abspath(os.path.join(fileDir, '..'))
sys.path.insert(0, dirPath)
from logBuilder.logger import AppLogger
updatelog = AppLogger()
class featureEng(object):
def processOutliers(self, dataFrame, continuous):
try:
overallIndeces = set()
outlierIndeces = {}
for feature in dataFrame[continuous].columns.values:
q1 = np.percentile(dataFrame[feature], 25)
q3 = np.percentile(dataFrame[feature], 75)
iqr = q3 - q1
outliers = dataFrame.loc[(dataFrame[feature] < q1 - 1.5 * iqr)
|
(dataFrame[feature] > q3 + 1.5 * iqr)]