-
Notifications
You must be signed in to change notification settings - Fork 0
/
DiversitySampling_Clustering.py
199 lines (150 loc) · 7.44 KB
/
DiversitySampling_Clustering.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
import os
from random import shuffle
import platform
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
from Cluster import CosineClusters
from Cluster import Cluster
from PreProcess import PCA_scale
if platform.system() == 'Windows':
wrk_path_win = r"C:\Users\Calvin\OneDrive\Documents\2022\Data_Output"
os.chdir(wrk_path_win)
elif platform.system() == 'Darwin':
wrk_path_3 = r"/Users/calvin/Documents/OneDrive/Documents/2022/Data_Output"
os.chdir(wrk_path_3)
'''
Pull in Unlabelled data
'''
# TODO need to determine how to implement it into a class (is it even required?)
def unlabelled_data(file, method):
ul_df = pd.read_csv(file)
column_drop = ['Duplicate_Check',
'PdI Width (d.nm)',
'PdI',
'Z-Average (d.nm)',
'ES_Aggregation']
ul_df = ul_df.drop(columns=column_drop)
ul_df.replace(np.nan, 'None', inplace=True)
ul_df = pd.get_dummies(ul_df, columns=["Component_1", "Component_2", "Component_3"],
prefix="", prefix_sep="")
# if method=='fillna': ul_df['Component_3'] = ul_df['Component_3'].apply(lambda x: None if pd.isnull(x) else x) #TODO This should be transformed into an IF function, thus when the function for unlabelled is filled with a parameter, then activates
ul_df = ul_df.groupby(level=0, axis=1, sort=False).sum()
print(ul_df.isna().any())
X_val = ul_df.to_numpy()
columns_x_val = ul_df.columns
return X_val, columns_x_val
X_val, columns_x_val = unlabelled_data('unlabelled_data_full.csv',
method='fillna') # TODO need to rename
class DiversitySampling():
def __init__(self, verbose=False):
self.verbose = verbose
self.pca = PCA_scale()
def get_cluster_samples(self, data, num_clusters=10, max_epochs=10, limit=-1):
if limit > 0:
data_new = list(zip(data[0], data[1]))
shuffle(data_new)
a, b = zip(*data_new)
a = np.array(a)
b = np.array(b)
data = tuple([a, b])
#shuffle(data)
data = data[:limit]
self.cosine_clusters = CosineClusters(num_clusters)
self.cosine_clusters.add_random_training_items(data[0], data[1])
for i in range(0, max_epochs):
print('Epoch ' + str(i))
added = self.cosine_clusters.add_items_to_best_cluster(data[0], data[1])
if added == 0:
break
self.centroids = self.cosine_clusters.get_centroids()
self.outliers = self.cosine_clusters.get_outliers(2) #Changed it to 2...
self.randoms = self.cosine_clusters.get_randoms(3)
return self.centroids, self.outliers, self.randoms
def get_representative_samples(self, training_data, unlabeled_data, number=20, limit=10000):
if limit > 0:
shuffle(training_data)
training_data = training_data[:limit]
shuffle(unlabeled_data)
unlabeled_data = unlabeled_data[:limit]
training_cluster = Cluster()
for index, item in enumerate(training_data):
training_cluster.add_to_cluster(item)
unlabeled_cluster = Cluster()
for index, item in enumerate(unlabeled_data):
unlabeled_cluster.add_to_cluster(item)
for index, item in enumerate(unlabeled_data):
training_score = training_cluster.cosine_similarity(item)
unlabeled_score = unlabeled_cluster.cosine_similarity(item)
representativeness = unlabeled_score - training_score
unlabeled_data.sort(reverse=True, key=lambda x: x[4])
return unlabeled_data[:number:]
def get_adaptive_representative_samples(self, training_data, unlabeled_data, number=20, limit=5000):
samples = []
for i in range(0, number):
print("Epoch " + str(i))
representative_item = self.get_representative_samples(training_data, unlabeled_data, 1, limit)[0]
samples.append(representative_item)
unlabeled_data.remove(representative_item)
return samples
def graph_clusters(self, data):
sample_y = [self.cosine_clusters.clusters.index(_) for _ in self.cosine_clusters.item_cluster.values()]
for index, cluster in enumerate(self.cosine_clusters.clusters):
sample_sort = cluster.distance_sort()
# Extract out the centroids/outliers/randoms
centroid_array = []
for _ in self.centroids:
for i in _:
centroid_array.append(i[2])
outlier_array = []
for _ in self.outliers:
for i in _:
outlier_array.append(i[2])
random_array = []
for _ in self.randoms:
for i in _:
random_array.append(i[2])
# PCA Fit & Transform Data - must be done on the fullscope of the data
self.pca.pca_fit(data[1])
data = self.pca.pca_transform(data[1])
# PCA Transform
centroid_array = self.pca.pca_transform(centroid_array)
outlier_array = self.pca.pca_transform(outlier_array)
random_array = self.pca.pca_transform(random_array)
D_id_color = [u'orchid', u'darkcyan', u'dodgerblue', u'turquoise', u'darkviolet', u'chartreuse', u'gold',
u'tomato', u'crimson', u'yellow', u'maroon', u'black', u'bisque', u'aqua', u'navy', u'magenta',
u'fuchsia', u'peru', u'red', u'brown', u'cornflowerblue']
plt.figure(figsize=(18, 6))
plt.subplot(131)
plt.scatter(data[:, 0], data[:, 1])
plt.subplot(132)
for label in [*range(len(self.cosine_clusters.clusters))]:
indices = [i for i, l in enumerate(sample_y) if l == label]
current_tx = np.take(data[:, 0], indices)
current_ty = np.take(data[:, 1], indices)
print(label)
color = D_id_color[label]
print(current_tx.shape)
plt.scatter(current_tx, current_ty, c=color, label=label)
plt.legend(bbox_to_anchor=(1.04,0), loc="lower left", borderaxespad=0)
plt.subplot(133)
plt.scatter(data[:, 0], data[:, 1], alpha=0.2, color='gray')
f2 = lambda x: [_[2] for _ in x]
for label in [*range(len(self.cosine_clusters.clusters))]:
color = D_id_color[label]
plt.scatter(np.take(centroid_array[label], 0), np.take(centroid_array[label], 1), c=color
, label=f'{label} centroids')
plt.scatter(np.take(outlier_array[label], 0), np.take(outlier_array[label], 1), marker ='*', c=color
, label=f'{label} outliers')
plt.scatter(np.take(random_array[label], 0), np.take(random_array[label], 1),marker='^', c=color
, label=f'{label} randoms')
#plt.scatter(np.array(f2(self.centroids[label]))[:, 0], np.array(f2(self.centroids[label]))[:, 1], c=color,
# label=f'{label} centroids')
#plt.scatter(np.array(f2(self.outliers[label]))[:, 0], np.array(f2(self.outliers[label]))[:, 1], marker='*',
# c=color,
# label=f'{label} outliers')
#plt.scatter(np.array(f2(self.randoms[label]))[:, 0], np.array(f2(self.randoms[label]))[:, 1], marker='^',
# c=color,
# label=f'{label} randoms')
plt.legend(bbox_to_anchor=(1.04, 0), loc="lower left", borderaxespad=0)
plt.show()