-
Notifications
You must be signed in to change notification settings - Fork 0
/
PreProcessing.py
209 lines (185 loc) · 6.64 KB
/
PreProcessing.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
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
__author__ = 'ghahr'
import csv
import json
import math
import random
from matplotlib import pyplot
import numpy as np
from sklearn.cluster import KMeans
from sklearn.decomposition import PCA
from sklearn import manifold
from sklearn.metrics import euclidean_distances
from sklearn.metrics.pairwise import cosine_similarity
from scipy.spatial.distance import correlation
def readData(fileName):
datasetDict = []
dataset = []
with open(fileName) as csvfile:
reader = csv.DictReader(csvfile)
print('here')
nameOfAttributes = reader.fieldnames
for row in reader:
datasetDict.append(row)
dataset.append(row.values())
# print row.values()
print('after')
return datasetDict, dataset, nameOfAttributes
def write_in_file(fileName, data, columnNames):
with open(fileName, 'w') as csvfile:
writer = csv.DictWriter(csvfile, fieldnames=columnNames)
writer.writeheader()
for d in data:
dict = {}
for i in range(len(d)):
dict[columnNames[i]] = d[i]
writer.writerow(dict)
def preprocess_data(data, nameOfAttributes):
newDataset = []
# print data[0]
avgValues = []
size1 = len(data)
size2 = len(data[0])
newAttributes = []
for i in range(size2):
sum = 0
count = 0
flag = False
for j in range(size1):
try:
if(data[j][i] != ''):
sum += float(data[j][i])
count += 1
except ValueError:
flag = True
break
if(flag == True or count == 0):
avgValues.append("unknown")
else:
avgValues.append(sum/count)
newAttributes.append(nameOfAttributes[i])
for d in data:
newD = []
for i in range(len(d)):
try:
if(avgValues[i] != "unknown"):
if(d[i] == ''):
newD.append(avgValues[i])
else:
newD.append(float(d[i]))
except ValueError:
pass
newDataset.append(newD)
return newDataset, newAttributes
def printData(printedData):
print(len(printedData))
for i in range(len(printedData)):
# print(printedData[i]('Year'))
print(json.dumps(printedData[i]))
def random_sampling(dataset, labels, centroids, sampledDataSize):
sampledData = []
sampledLabels = []
for i in range(sampledDataSize):
index = random.randint(0, len(dataset)-1)
sampledData.append(dataset[index])
sampledLabels.append(labels[index])
return sampledData, sampledLabels
def write_kmean_result_in_file(data):
centroids, labels = k_mean_clustering(data)
with open("kmean_result.csv", 'wb') as csvfile:
spamwriter = csv.writer(csvfile)
spamwriter.writerow(labels)
for c in centroids:
spamwriter.writerow(c)
def read_k_mean_clustering():
with open("kmean_result.csv") as csvfile:
spamreader = csv.reader(csvfile)
list_rows = []
for row in spamreader:
list_rows.append(row)
return list_rows[0], list_rows[1:]
def adaptive_sampling(data, labels, centroids, sampledDataSize):
for i in range(len(labels)):
labels[i] = int(labels[i])
probabilities = [0] * len(centroids)
for l in labels:
probabilities[l] += 1
for i in range(len(probabilities)):
probabilities[i] = probabilities[i] * 1.0 / len(labels)
sampledData = []
sampledLabels = []
i = 0
while(i<sampledDataSize):
index = random.randint(0, len(data) - 1)
if(random.random >= probabilities[labels[index]]):
sampledData.append(data[index])
sampledLabels.append(labels[index])
i += 1
return sampledData, sampledLabels
def k_mean_clustering(data):
k = 8
k_range = range(1,14)
newData = np.array(data)
# centroids,labels,inertia = cluster.k_means(newData,n_clusters=k)
kmeans_var = KMeans(init='k-means++', n_clusters=k, n_init=10).fit(newData)
centroids = kmeans_var.cluster_centers_
labels = kmeans_var.labels_
return centroids, labels
# http://scikit-learn.org/stable/auto_examples/decomposition/plot_pca_3d.html
def my_pca(data):
X = np.array(data)
pca = PCA(n_components=2)
pca.fit(X)
results2 = pca.transform(X)
return results2
def find_intrinsic_dimensionality(data):
X = np.array(data)
pca = PCA(n_components=10)
pca.fit(X)
results = pca.explained_variance_ratio_
return results
def split_array(myarray):
X = []
Y = []
for el in myarray:
X.append(el[0])
Y.append(el[1])
return X, Y
def euclidean_MDS(data):
seed = np.random.RandomState(seed=3)
similarities = euclidean_distances(data)
mds = manifold.MDS(n_components=2, max_iter=3000, eps=1e-9, random_state=seed,
dissimilarity="precomputed", n_jobs=1)
pos = mds.fit_transform(similarities)
return pos
def cosine_MDS(data):
seed = np.random.RandomState(seed=3)
similarities = cosine_similarity(data, data)
mds = manifold.MDS(n_components=2, max_iter=3000, eps=1e-9, random_state=seed,
dissimilarity="precomputed", n_jobs=4)
pos = mds.fit_transform(similarities)
return pos
# http://docs.scipy.org/doc/scipy-0.14.0/reference/generated/scipy.spatial.distance.correlation.html
def correlation_MDS(data):
seed = np.random.RandomState(seed=3)
similarities = [[0 for x in range(len(data))] for x in range(len(data))]
for i in range(len(data)):
for j in range(len(data)):
similarities[i][j] = correlation(data[i], data[j])
mds = manifold.MDS(n_components=2, max_iter=3000, eps=1e-9, random_state=seed,
dissimilarity="precomputed", n_jobs=1)
pos = mds.fit_transform(similarities)
return pos
def isomap(data):
X = np.array(data)
X_iso = manifold.Isomap(n_neighbors = 10, n_components=2).fit_transform(X)
return X_iso
if __name__ == "__main__":
# at first, I read data and preprocessed them and then write processed data in a new file
# datasetDict, original_dataset, original_nameOfAttributes = readData('prosperLoanData.csv')
# dataset, nameOfAttributes = preprocess_data(original_dataset, original_nameOfAttributes)
# write_in_file("processed_data.csv", dataset, nameOfAttributes)
datasetDict, original_dataset, original_nameOfAttributes = readData('processed_data.csv')
# sampledData, sampledLabels = adaptive_sampling(dataset)
# # pcaData = my_pca(sampledData, nameOfAttributes)
# print "before"
# correlation_MDS(sampledData)