-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathtraining_ver7_System1_1.py
161 lines (127 loc) · 7.76 KB
/
training_ver7_System1_1.py
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
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
# -*- coding: utf-8 -*-
"""
Created on Thu Jul 2 21:31:23 2020
@author: bcosk
"""
import pandas as pd
import numpy as np
from tensorflow import keras
from keras.models import Sequential
from keras.layers import Dense
from keras import optimizers
from keras.objectives import mean_squared_error, mean_absolute_error
from keras.layers import LSTM, Flatten
from keras.callbacks import EarlyStopping
import matplotlib.pyplot as plt
from datetime import datetime
from sklearn.tree import DecisionTreeRegressor
def dataPreparing(dataFirst,colNumber):
newColumnNames = []
for i in range(0,len(dataFirst.columns)):
newColumnNames.append(dataFirst.columns[i].replace("T","Q"))
confidenceFirst = pd.DataFrame(np.zeros((len(dataFirst),len(dataFirst.columns))), columns = newColumnNames)
inputData = dataFirst.to_numpy()
outputData = confidenceFirst.to_numpy()
outputData1 = dataFirst.to_numpy()
processedDataFirst = dataFirst.copy()
processedconfidenceFirst = confidenceFirst.copy()
for i in range(0,len(processedDataFirst)):
tempConfidence = (i%201 - 100)/100
processedDataFirst[processedDataFirst.columns[colNumber]][i] += processedDataFirst[processedDataFirst.columns[colNumber]][i] * tempConfidence
processedconfidenceFirst[processedconfidenceFirst.columns[colNumber]][i] += tempConfidence
return processedDataFirst, processedconfidenceFirst[processedconfidenceFirst.columns[colNumber]], dataFirst[dataFirst.columns[colNumber]]
dataFirst = pd.read_csv("P:\Tez\Data\ER12.csv", sep = ";")
dataTest = pd.read_csv("P:\Tez\Data\ER13.csv", sep = ";")
errorList0 = []
errorList1 = []
now = datetime.now()
for selectedData in range(0,len(dataTest.columns)):
# for selectedData in range(9,10):
inputData, outputData, outputData1 = dataPreparing(dataFirst,selectedData)
inputTestData, outputTestData, outputTestData1 = dataPreparing(dataTest,selectedData)
epochs = 1000
validation_split = 0.2
verbose = 1 # 0 silent, 1 progress, 2 text
shuffle = True
patience = 3
mape = keras.losses.MeanAbsolutePercentageError(name='mape')
es = EarlyStopping(monitor="val_mae", mode='auto', verbose=verbose, patience=patience, restore_best_weights=True)
model = Sequential()
model.add(Dense(22,activation="relu",input_shape=(11,)))
model.add(Dense(1))
model.compile(optimizer="adam",loss = "mae", metrics=["mae"])
history = model.fit(inputData,outputData,epochs=epochs,validation_split=validation_split, verbose = verbose, shuffle = shuffle, callbacks=[es])
testResult = model.evaluate(inputTestData,outputTestData,verbose=verbose)
max_depth=10
rf = DecisionTreeRegressor(criterion = "mae",random_state = 0,max_depth=max_depth)
rf.fit(inputData, outputData)
if selectedData == 8:
newColumnNames = []
for i in range(0,len(dataTest.columns)):
newColumnNames.append(dataTest.columns[i].replace("T","Q"))
confidenceFirst = pd.DataFrame(np.zeros((len(dataTest),len(dataTest.columns))), columns = newColumnNames)
processedDataFirst = dataTest.copy()
processedconfidenceFirst = confidenceFirst.copy()
tempConfidence = 0.5
for i in range(int(len(dataTest)/2),len(dataTest)):
processedDataFirst[processedDataFirst.columns[selectedData]][i] += processedDataFirst[processedDataFirst.columns[selectedData]][i] * tempConfidence
processedconfidenceFirst[processedconfidenceFirst.columns[selectedData]][i] += tempConfidence
dataPredicted = dataTest.copy()
confidencePredicted = confidenceFirst.copy()
confidencePredicted[confidencePredicted.columns[selectedData]] = model.predict(processedDataFirst)
confidencePredicted1 = confidenceFirst.copy()
confidencePredicted1[confidencePredicted1.columns[selectedData]] = rf.predict(processedDataFirst)
rangeMin = 40000
rangeMax = 80000
ax = processedconfidenceFirst.reset_index().iloc[rangeMin:rangeMax].plot.scatter(y = processedconfidenceFirst.columns[selectedData],x="index",s=0.05,c="red")
confidencePredicted.reset_index().iloc[rangeMin:rangeMax].plot.scatter(y = confidencePredicted.columns[selectedData],x="index",s=0.05,c="blue",ax=ax)
plt.legend(["v","Predicted v by Neural Network"])
plt.grid()
plt.xlabel("Timestamp")
plt.ylabel("Noise Ratio")
plt.savefig(now.strftime("%d_%m_%Y_%H_%M_%S") + "System1_1", dpi = 300)
plt.show()
ax = processedconfidenceFirst.reset_index().iloc[rangeMin:rangeMax].plot.scatter(y = processedconfidenceFirst.columns[selectedData],x="index",s=0.05,c="red")
confidencePredicted1.reset_index().iloc[rangeMin:rangeMax].plot.scatter(y = confidencePredicted1.columns[selectedData],x="index",s=0.05,c="blue",ax=ax)
plt.legend(["v","Predicted v by Decision Tree"])
plt.grid()
plt.xlabel("Timestamp")
plt.ylabel("Noise Ratio")
plt.savefig(now.strftime("%d_%m_%Y_%H_%M_%S") + "System1_2", dpi = 300)
plt.show()
print("Testing: " + str(selectedData))
tempErrorList0 = []
tempErrorList1 = []
tempErrorList2 = []
for k in range(-10,11):
newColumnNames = []
for i in range(0,len(dataTest.columns)):
newColumnNames.append(dataTest.columns[i].replace("T","Q"))
confidenceFirst = pd.DataFrame(np.zeros((len(dataTest),len(dataTest.columns))), columns = newColumnNames)
processedDataFirst = dataTest.copy()
processedconfidenceFirst = confidenceFirst.copy()
tempConfidence = k/10
for i in range(0,len(dataTest)):
processedDataFirst[processedDataFirst.columns[selectedData]][i] += processedDataFirst[processedDataFirst.columns[selectedData]][i] * tempConfidence
processedconfidenceFirst[processedconfidenceFirst.columns[selectedData]][i] += tempConfidence
dataPredicted = dataTest.copy()
confidencePredicted = confidenceFirst.copy()
confidencePredicted[confidencePredicted.columns[selectedData]] = model.predict(processedDataFirst)
# dataPredicted[dataPredicted.columns[selectedData]] = processedDataFirst[processedDataFirst.columns[selectedData]] - (processedDataFirst[processedDataFirst.columns[selectedData]] * confidencePredicted[confidencePredicted.columns[selectedData]])
dataPredicted1 = dataTest.copy()
confidencePredicted1 = confidenceFirst.copy()
confidencePredicted1[confidencePredicted1.columns[selectedData]] = rf.predict(processedDataFirst)
# dataPredicted1[dataPredicted1.columns[selectedData]] = processedDataFirst[processedDataFirst.columns[selectedData]] - (processedDataFirst[processedDataFirst.columns[selectedData]] * confidencePredicted1[confidencePredicted1.columns[selectedData]])
errorList = abs((confidencePredicted[confidencePredicted.columns[selectedData]] - processedconfidenceFirst[processedconfidenceFirst.columns[selectedData]]))
tempErrorList0.append(sum(errorList)/len(errorList))
errorList = abs((confidencePredicted1[confidencePredicted1.columns[selectedData]] - processedconfidenceFirst[processedconfidenceFirst.columns[selectedData]]))
tempErrorList1.append(sum(errorList)/len(errorList))
errorList0.append(tempErrorList0)
errorList1.append(tempErrorList1)
dfErrorList0 = pd.DataFrame(errorList0)
dfErrorList1 = pd.DataFrame(errorList1)
resultExcelName = now.strftime("%d_%m_%Y_%H_%M_%S") + "_System1_Testing.xlsx"
writer = pd.ExcelWriter(resultExcelName, engine='xlsxwriter')
dfErrorList0.to_excel(writer, sheet_name='NeuralNetwork', index=False)
dfErrorList1.to_excel(writer, sheet_name='DecisionTree', index=False)
writer.save()