CNN-for-PD-Action/FCN.py at master · rohithkg-code/CNN-for-PD-Action · GitHub

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
# -*- coding: utf-8 -*-
"""
Created on 2021

"""

from __future__ import print_function
from tensorflow import keras
from sklearn.metrics import classification_report
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import xlwt

from sklearn.metrics import roc_auc_score

def readucr(filename):
    data = np.loadtxt(filename, delimiter = ',')
    Y = data[:,0]
    X = data[:,1:]
    return X, Y

nb_epochs = 1000

#flist = ['Adiac', 'Beef', 'CBF', 'ChlorineConcentration', 'CinC_ECG_torso', 'Coffee', 'Cricket_X', 'Cricket_Y', 'Cricket_Z',
#'DiatomSizeReduction', 'ECGFiveDays', 'FaceAll', 'FaceFour', 'FacesUCR', '50words', 'FISH', 'Gun_Point', 'Haptics',
#'InlineSkate', 'ItalyPowerDemand', 'Lighting2', 'Lighting7', 'MALLAT', 'MedicalImages', 'MoteStrain', 'NonInvasiveFatalECG_Thorax1',
#'NonInvasiveFatalECG_Thorax2', 'OliveOil', 'OSULeaf', 'SonyAIBORobotSurface', 'SonyAIBORobotSurfaceII', 'StarLightCurves', 'SwedishLeaf', 'Symbols',
#'synthetic_control', 'Trace', 'TwoLeadECG', 'Two_Patterns', 'uWaveGestureLibrary_X', 'uWaveGestureLibrary_Y', 'uWaveGestureLibrary_Z', 'wafer', 'WordsSynonyms', 'yoga']

flist  = ['PD180']
for each in flist:
    fname = each
    x_train, y_train = readucr(fname+'/'+fname+'_TRAIN')
    x_test, y_test = readucr(fname+'/'+fname+'_TEST')
    nb_classes = len(np.unique(y_test))
    batch_size = 10 # min(x_train.shape[0]/10, 16)

    #y_train = (y_train - y_train.min())/(y_train.max()-y_train.min())*(nb_classes-1)
    #y_test = (y_test - y_test.min())/(y_test.max()-y_test.min())*(nb_classes-1)

    Y_train = keras.utils.to_categorical(y_train, nb_classes)
    Y_test = keras.utils.to_categorical(y_test, nb_classes)

    x_train_mean = x_train.mean()
    x_train_std = x_train.std()
    x_train = (x_train - x_train_mean)/(x_train_std)

    x_test = (x_test - x_train_mean)/(x_train_std)
    x_train = x_train.reshape(x_train.shape + (1,1,))
    x_test = x_test.reshape(x_test.shape + (1,1,))

    x = keras.layers.Input(x_train.shape[1:])
#    drop_out = Dropout(0.2)(x)
    conv1 = keras.layers.Conv2D(64, 8, 1, padding='same')(x)
    conv1 = keras.layers.BatchNormalization()(conv1)
    conv1 = keras.layers.Activation('relu')(conv1)

#    drop_out = Dropout(0.2)(conv1)
    conv2 = keras.layers.Conv2D(128, 5, 1, padding='same')(conv1)
    conv2 = keras.layers.BatchNormalization()(conv2)
    conv2 = keras.layers.Activation('relu')(conv2)

#    drop_out = Dropout(0.2)(conv2)
    conv3 = keras.layers.Conv2D(64, 3, 1, padding='same')(conv2)
    conv3 = keras.layers.BatchNormalization()(conv3)
    conv3 = keras.layers.Activation('relu')(conv3)

    full = keras.layers.GlobalAveragePooling2D()(conv3)
    out = keras.layers.Dense(nb_classes, activation='softmax')(full)


    model = keras.models.Model(inputs=x, outputs=out)

    optimizer = keras.optimizers.Adam()
    model.compile(loss='categorical_crossentropy',
                  optimizer=optimizer,
                  metrics=['accuracy'])

    reduce_lr = keras.callbacks.ReduceLROnPlateau(monitor = 'loss', factor=0.5,
                      patience=50, min_lr=0.0001)
    hist = model.fit(x_train, Y_train, batch_size=batch_size, epochs=nb_epochs,
              verbose=1,
              validation_data=(x_test, Y_test),
              callbacks = [reduce_lr])
    #Print the testing results which has the lowest training loss.
    log = pd.DataFrame(hist.history)
    print(model.summary())
    #flops = get_flops(model, batch_size=16)
    #print(f"FLOPS: {flops / 10 ** 9:.03} G")


    # 创建一个workbook 设置编码
    workbook = xlwt.Workbook(encoding = 'utf-8')
    # 创建一个worksheet
    worksheet = workbook.add_sheet('My Worksheet')
    print(Y_test.shape[0])
    num = Y_test.shape[0]
    y_pred = model.predict(x_test)

    auc_score = roc_auc_score(Y_test,y_pred)
    print(auc_score)

    for i in range(len(y_pred)):
      max_value=max(y_pred[i])
      for j in range(len(y_pred[i])):
        if max_value==y_pred[i][j]:
          y_pred[i][j]=1
        else:
          y_pred[i][j]=0
    print(classification_report(Y_test, y_pred))


    for i  in range(0,1):
        y_row=Y_test[i]
        p_row=y_pred[i]
        #print(y_row)
        #print(p_row)
        print(np.argmax(y_row))
        print(np.argmax(p_row))
        worksheet.write(i,0,str(np.argmax(y_row)))
        worksheet.write(i,1,str(np.argmax(p_row)))
    # 保存
    workbook.save('test.xls')
    print(log.loc[log['loss'].idxmin]['loss'], log.loc[log['loss'].idxmin]['val_acc'])

epochs=range(len(hist.history['acc']))
plt.figure()
plt.plot(epochs,hist.history['acc'],'b',label='Training acc')
plt.plot(epochs,hist.history['val_acc'],'r',label='Validation acc')
plt.title('Traing and Validation accuracy')
plt.legend()
plt.savefig('1_acc.jpg')

plt.figure()
plt.plot(epochs,hist.history['loss'],'b',label='Training loss')
plt.plot(epochs,hist.history['val_loss'],'r',label='Validation val_loss')
plt.title('Traing and Validation loss')
plt.legend()
plt.savefig('1_loss.jpg')