-
Notifications
You must be signed in to change notification settings - Fork 0
/
Figure6.py
235 lines (202 loc) · 9.32 KB
/
Figure6.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
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
""" 比率变换对于识别精度的变化, 已经改成了lot的交叉验证的形式"""
import numpy as np
from sklearn.decomposition import PCA
import pandas as pd
from User_modify import *
from Tool_Dataprocess import produce_sample_label
import matplotlib.pyplot as plt
from sklearn import preprocessing
from sklearn.neighbors import KNeighborsClassifier
from sklearn import svm
from sklearn.ensemble import RandomForestClassifier
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
from sklearn.neural_network import MLPClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn.model_selection import StratifiedKFold
import sklearn.metrics as sm
import numpy as np
from Tool_Visualization import *
from sklearn.model_selection import KFold
from sklearn.model_selection import cross_val_score
from sklearn.model_selection import GridSearchCV
from sklearn.model_selection import train_test_split
from sklearn.model_selection import LeaveOneGroupOut
from Tool_Functionset import *
from Tool_Functionset import *
def multi_set_size(dataset, label, candidate_model_a, candidate_model_b, candidate_model_C, flag, frac):
"""
目的是: 为model_a，和model_b选择不同的分类器，验证在lot验证下的分类精度
:param dataset:
:param label:
:return: force_score_lot_array; gesture_score_lot_array; total_score_lot_array
"""
gesture_score_list = [] # 记录每一次交叉验证的手势分类精度
force_score_list = [] #记录每一次力分类的精度
total_score_list = [] #记录总共的分类精度
gesture_recall_list = []
force_recall_list = [] #记录每一次力分类的精度
total_recall_list = [] #记录总共的分类精度
fold = 10
for i in range(fold):
j = i
test_index = label[:, 3] == i
test_index_1 = label[:, 3] == j
for m in range(test_index_1.shape[0]):
test_index[m] = test_index[m] or test_index_1[m]
train_index = ~ test_index
train_dataset = dataset[train_index, :]
train_label = label[train_index, :]
test_dataset = dataset[test_index, :]
test_label = label[test_index, :]
batch_size = int(np.floor(frac * train_dataset.shape[0]))
train_dataset = train_dataset[0:batch_size,:]
train_label = train_label[0:batch_size,:]
#建立3种分类的模型
model_a = candidate_model_a
model_b = candidate_model_b
model_c = candidate_model_C
# 进行归一化
scaler = preprocessing.MinMaxScaler(feature_range=(0, 1)).fit(train_dataset)
train_dataset = scaler.transform(train_dataset)
# 对test_dataset进行归一化
test_dataset = scaler.transform(test_dataset)
"""手势分类 """
# 设置model a 的标签
train_label_a = train_label[:, 1]
test_label_a = test_label[:, 1]
model_a.fit(train_dataset, train_label_a)
#
predict_a = model_a.predict(test_dataset) # 给出第一次预测的结果
test_score_a = model_a.score(test_dataset, test_label_a)
test_recall_a = sm.recall_score(y_true=test_label_a, y_pred=predict_a, average='macro')
gesture_score_list.append(test_score_a) # 把lot交叉验证的分数记录下来
gesture_recall_list.append(test_recall_a)
""" 力分类"""
# 设置model b 的标签
train_label_b = train_label[:, 2]
test_label_b = test_label[:, 2]
model_b.fit(train_dataset, train_label_b)
#
predict_b = model_b.predict(test_dataset) # 给出第一次预测的结果
test_score_b = model_b.score(test_dataset, test_label_b)
force_score_list.append(test_score_b) # 把lot交叉验证的分数记录下来
test_recall_b = sm.recall_score(y_true=test_label_b, y_pred=predict_b, average='macro')
force_recall_list.append(test_recall_b)
""" 总分类 """
#设置model_c的标签
train_label_c = train_label[:, 4]
test_label_c = test_label[:, 4]
model_c.fit(train_dataset, train_label_c)
#
predict_c = model_c.predict(test_dataset) # 给出第一次预测的结果
test_score_c = model_c.score(test_dataset, test_label_c)
total_score_list.append(test_score_c) # 把lot交叉验证的分数记录下来
test_recall_c = sm.recall_score(y_true=test_label_c, y_pred=predict_c, average='macro')
total_recall_list.append(test_recall_c)
gesture_score = np.mean(gesture_score_list)
force_score = np.mean(force_score_list)
total_score = np.mean(total_score_list)
gesture_recall = np.mean(gesture_recall_list)
force_recall = np.mean(force_recall_list)
total_recall = np.mean(total_recall_list)
if flag ==1:
return gesture_score, force_score, total_score
else:
return gesture_recall, force_recall, total_recall
sub_num = 14
# rate_score_array = []
# for sub_index in range(1,sub_num+1):
# path1 = './Feature_2/data' + str(sub_index) + '3.csv'
# path2 = './Feature_2/label' + str(sub_index) + '3.csv'
#
# dataset = pd.read_csv(path1, header=None)
# label = pd.read_csv(path2, header=None)
# dataset = np.array(dataset)
# label = np.array(label)
# label = label.astype(int)
#
# # 打乱数据，为各个训练器准备训练数据
# data_pool = np.zeros([dataset.shape[0], dataset.shape[1] + label.shape[1]])
# data_pool[:, 0:dataset.shape[1]] = dataset
# data_pool[:, dataset.shape[1]:] = label
# np.random.seed(0) # 控制随机数产生的种子
# np.random.shuffle(data_pool)
# dataset = data_pool[:, 0:dataset.shape[1]]
# label = data_pool[:, dataset.shape[1]:]
# label = label.astype(int)
#
# # 构造分类器的集合
# knn = KNeighborsClassifier() # KNN Modeling
# clf = svm.SVC(kernel='rbf', C=10)
# rf = RandomForestClassifier(n_estimators=16) # RF Modeling
# lda = LinearDiscriminantAnalysis()
# selected_model = [rf, lda, knn, clf]
# force_score_classifier_array = np.zeros((4, 10))
# gesture_score_classifier_array = np.zeros((4, 10))
#
# # 循环选择模型
# model_score_array = []
# output_flag = 1
# frac_list = [0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1]
# for frac in frac_list:
# #根据预实验的结果，分类器选择SVM分类器
# candidate_model_a = selected_model[3] # 选择只针对手势分类的分类器
# candidate_model_b = selected_model[3] # 选择只针对力分类的分类器
# candidate_model_c = selected_model[3] # 选择总共的分类器
#
#
# gesture_score, force_score, total_score = multi_set_size(dataset, label, candidate_model_a, candidate_model_b,
# candidate_model_c, flag=output_flag, frac=frac)
# print(frac)
# score_list = [gesture_score, force_score, total_score]
# model_score_array.append(score_list)
# model_score_array = np.array(model_score_array)
# rate_score_array.append(model_score_array)
# print(sub_index)
#
# rate_score_array = np.array(rate_score_array)#0维：subject 1维度:frac 2维：3个score
# np.save('./new_figure/rate_score_array.npy', rate_score_array)
# print(1)
#
rate_score_array = np.load('./new_figure/rate_score_array.npy')
linestyle_list = ['-','--','-.']
marker_list = ['o','*','1']
color_list = ['skyblue', 'lightcoral', 'navajowhite']
legend_list = ['Hand gesture classification', 'Force level classification', 'Simultaneous classification of hand gestures and force levels ']
xlabel = ['10%', '20%', '30%','40%','50%','60%','70%','80%','90%','100%']
RA_flag =1
if RA_flag ==1:
data = np.mean(rate_score_array, axis=0)
data_max = np.max(rate_score_array, axis=0)
data_min = np.min(rate_score_array, axis=0)
data_std = np.std(rate_score_array, axis=0)
print(1)
plt.rcParams['font.family'] = 'Arial'
plt.rcParams.update({'font.size': 6})
plt.figure(figsize=(3.5, 3), dpi=800)
ax = plt.gca()
ax.spines['top'].set_linewidth(0)
ax.spines['right'].set_linewidth(0)
bwith = 0.5
ax.spines['bottom'].set_linewidth(bwith)
ax.spines['left'].set_linewidth(bwith)
x = np.arange(data.shape[0])
for i in range(3):
plt.scatter(x, data[:, i] * 100, marker=marker_list[i], s=5)
plt.plot(x, data[:,i]*100,label = legend_list[i], color= color_list[i], linestyle=linestyle_list[i] )
# plt.fill_between(x, (data[:,i]+data_std[:,i])*100, (data[:,i]-data_std[:,i])*100, # 上限，下限
# facecolor=color_list[i], # 填充颜色
# alpha=0.3) # 透明度
plt.ylim([60,90])
plt.xticks(x,xlabel,fontsize = 6)
plt.tick_params(width=0.5) # 设置刻度线条的粗细
plt.grid(axis='y')
plt.ylabel('Recognition Accuracy(%)', fontsize = 7)
plt.xlabel('Percentage of Training set', fontsize = 7)
#plt.legend(loc = 'lower right',fontsize = 6)
leg = plt.legend(loc = 'upper left',bbox_to_anchor = (-0.01,-0.25) ,fontsize = 7, ncol = 1)
leg.get_frame().set_linewidth(0.1)
plt.tight_layout()
#plt.savefig('C:\\Users\\86156\\Desktop\\f6_1.png')
plt.savefig('C:\\Users\\86156\\Desktop\\why\\f6_new_2.pdf',bbox_inches='tight', pad_inches=0.1)
#plt.savefig('C:\\Users\\Lenovo\\Desktop\\fig\\f6_new_2.png',bbox_inches='tight', pad_inches=0.1)#,bbox_inches='tight'