import ts_set
import csv
# onehotencoser is python file which onehotencode each event
# and return them alnog with their remaining time
import onehotencoder
# mydatetime is python file which return diffrence between
# two timestamp in seconds.
import mydatetime
# encodedlist is a list which stores encoded list of each event
encodedlist = onehotencoder.encodedlist()
states, events, transitions = ts_set.TransitionSystem()
# read the data of log.csv and stores it in csvlog which
# is list of lists where each list contain all the feature
# of the event
csvlog = open("../../../Dataset/log.csv", "r").read().split("\n")
event_number = 0
for line in csvlog:
line = line.split(",")
csvlog[event_number] = line
event_number += 1
noofrow = len(csvlog) - 1 #number of row in the dataset
noofcol = len(csvlog[0]) #number of coloumn in dataset
feature_data = []
import csv
import mydatetime
A = []
P = []
csvlog = open("../Dataset/log.csv", "r").read().split("\n")
j = 0
for i in csvlog:
i = i.split(",")
csvlog[j] = i
j += 1
noofrow = len(csvlog) - 1
k = noofrow
noofcol = len(csvlog[0])
a4 = 0
states, events, transitions, avgremtime = ts_set.TransitionSystem()
for i in range(1, noofrow // 5):
for j in range(i, noofrow + 1):
if csvlog[i][0] != csvlog[j][0]:
j = j - 1
break
if csvlog[j][3] != "Send for Credit Collection":
continue
curr_state = ['start']
if csvlog[i][0] != csvlog[i - 1][0]:
trace_start = i
for ii in range(trace_start, i + 1):
curr_state.append(csvlog[ii][3])
curr_state = list(set(curr_state))
for j in range(i, noofrow + 1):
if (csvlog[j][0] != csvlog[i][0]):