-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathclass_DistanceMetric.py
187 lines (138 loc) · 5.11 KB
/
class_DistanceMetric.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
import numpy as np
import os
import pandas as pd
import matplotlib.pyplot as plt
import collections
def open_data(datapath,file_type):
file_list = []
path_list = os.listdir(datapath)
for filename in path_list:
if file_type in filename:
file_list.append(os.path.join(datapath, filename))
return file_list
def read_data(file_name):
df1 = pd.read_csv(file_name, usecols = ['movement_label','umap1', 'umap2', 'zs_velocity'])
# print(df1,type(df1))
return df1
# def collect_data(data_list):
#
#
# movement_label_list = data_list.values[0:len(data_list), 0]
# create_label_list = [x*x for x in range(1, 4, 1)]
#
#
# for value in movement_label_list:
# if value in create_label_list:
# print(movement_label_list.index)
#
# return
def groupby_data(result):
merge_data = []
for i in range(0, len(result), 1):
merge_data.append(result[i])
merged_df = pd.concat(merge_data, ignore_index=False)
return merged_df
def mergedf_center(merged_data):
umap1_mean = merged_data['umap1'].mean()
umap2_mean = merged_data['umap2'].mean()
v_mean = merged_data['zs_velocity'].mean()
center = [umap1_mean, umap2_mean, v_mean]
return center
def data3D(merged_data):
umap1 = merged_data['umap1']
umap2 = merged_data['umap2']
v = merged_data['zs_velocity']
return umap1,umap2,v
def cluster_dis(center,x,y,z):
distance = [np.sqrt((x - center[0]) ** 2 + (y - center[1]) ** 2 + (z - center[2])** 2)]
return distance
def same_cluster_dis(file_list, label_num):
result = []
for i in range(0, len(file_list), 1):
data_list = read_data(file_list[i])
# group_data = data_list.groupby("movement_label")
# print(group_data)
# for key, df in group_data:
# print(key)
# print(df)
res = dict(tuple(data_list.groupby('movement_label')))
# result_ = 'result_' + str(i+1)
# result = result.append([res[i]])
result.append(res[label_num])
merged_df2 = groupby_data(result)
center = mergedf_center(merged_df2)
[umap1, umap2, v] = data3D(merged_df2)
distance = cluster_dis(center, umap1, umap2, v)
distance =distance[0].mean()
return distance
def diff_cluster_dis(file_list, label_num1, label_num2):
result1 = []
result2 = []
for i in range(0, len(file_list), 1):
data_list = read_data(file_list[i])
# group_data = data_list.groupby("movement_label")
# print(group_data)
# for key, df in group_data:
# print(key)
# print(df)
res = dict(tuple(data_list.groupby('movement_label')))
# result_ = 'result_' + str(i+1)
# result = result.append([res[i]])
result1.append(res[label_num1])
result2.append(res[label_num2])
merged_df1 = groupby_data(result1) #label_1 cluster
merged_df2 = groupby_data(result2) #label_2 cluster
center = mergedf_center(merged_df1) #label_1 center
[umap1, umap2, v] = data3D(merged_df2) #label_1 data
distance = cluster_dis(center, umap1, umap2, v)
distance = distance[0].mean()
return distance
if __name__ == '__main__':
file_list = open_data('D:/3D_behavior/looming_behavior/results-YJL/BeAMapping', 'Feature_Space.csv')
# print(file_list)
result = []
print(file_list)
for i in range(0, len(file_list), 1):
data_list = read_data(file_list[i])
# group_data = data_list.groupby("movement_label")
# print(group_data)
# for key, df in group_data:
# print(key)
# print(df)
res = dict(tuple(data_list.groupby('movement_label')))
# result_ = 'result_' + str(i+1)
# result = result.append([res[i]])
result.append(res)
# result -> list[dic1, dic2, .., dic12]
# each dic{1:dataframe, 2:dataframe, ..., 40:dataframe}
# merge to a big dict{1:dataframe, ..., 40:dataframe}
big_dict = {}
for item in result:
for behave in item: # behave key value is int
# 如果当前的行为不在大的big_dict, 那么我们就把这个行为放进去
# {行为：dataframe} 行为：1-40key
# item{1:dataframe, 2:dataframe}
if behave not in big_dict:big_dict.update({behave:item[behave]})
else:
big_dict[behave] = pd.concat([big_dict[behave], item[behave]])
print(len(big_dict))
big_dict_order = collections.OrderedDict(sorted(big_dict.items()))
# merged_df2 = groupby_data(result)
# center = mergedf_center(merged_df2)
# [umap1,umap2,v] = data3D(merged_df2)
# distance = cluster_dis(center,umap1,umap2,v)
# print(distance[0].mean())
# # plt.plot(distance[0],x=2)
# # plt.show()
# try:
#
# for j in range(1,40,1):
# label = same_cluster_dis(file_list, j)
# print(label)
#
# except:
# print(f'label{j} error'.format(j))
# j = j+1
# label = same_cluster_dis(file_list, j)
# print(label)
# diff_distance= diff_cluster_dis(file_list, 11, 11)