Question

目前我正在尝试为数据集csv中的故障检测构建深度稀疏去噪自动编码器我使用https://blog.keras.io/building-autoencoders-in-keras.html将其应用于数值数据集中。我的数据集是：train

和我的代码：

    import numpy as np
from keras.layers import Input, Dense
from keras.models import Model
from keras.datasets import mnist
import matplotlib.pyplot as plt
from keras import regularizers
import csv as csv
x_train=[]
x_test=[]         #Array Definition
path1 =  r'D:\train.csv'     #Address Definition
path2 =  r'D:\train.csv'
with open(path1, 'r') as f1:    #Open File as read by 'r'
    reader = csv.reader(f1)     
    next(reader, None)          #Skip header because file header is not needed
    for row in reader:          #fill array by file info by for loop
        x_train.append(row)
    x_train = np.array(x_train)         

with open(path2, 'r') as f2:
    reader2 = csv.reader(f2)
    next(reader2, None)  
    for row2 in  reader2:
        x_test.append(row2)
    x_test = np.array(x_test)

x_train = np.delete(x_train,[0],1)  #delete first column of which is patientid
x_test = np.delete(x_test,[0],1)



x_train = np.delete(x_train,[0],1) #Delete name column because it is not needed
x_test = np.delete(x_test,[0],1)

x_train = np.delete(x_train,[12],1) #Delete name column because it is not needed
x_test = np.delete(x_test,[12],1)

# this is our input placeholder
input_img = Input(shape=(12,))
# "encoded" is the encoded representation of the input
encoded = Dense(64, activation='relu',activity_regularizer=regularizers.l1(10e-6))(input_img)
encoded = Dense(128, activation='relu')(encoded)

# "decoded" is the lossy reconstruction of the input
decoded = Dense(128, activation='relu')(encoded)
decoded = Dense(64, activation='relu')(decoded)
decoded = Dense(12, activation='sigmoid')(decoded)
# this model maps an input to its reconstruction
autoencoder = Model(input_img, decoded)
# configure our model to use a per-pixel binary crossentropy loss, and the Adadelta optimizer:
autoencoder.compile(optimizer='adadelta', loss='binary_crossentropy')

# prepare our input data. We're using MNIST digits, and we're discarding the labels
#(x_train, _), (x_test, _) = mnist.load_data()
# normalize all values between 0 and 1 and we will flatten the 28x28 images into vectors of size 784.
x_train = x_train.astype('float32')/255.
x_test = x_test.astype('float32')/255.
#x_train = x_train.reshape((len(x_train), np.prod(x_train.shape[1:])))
#x_test = x_test.reshape((len(x_test), np.prod(x_test.shape[1:])))
noise_factor = 0.4
x_train_noisy = x_train + noise_factor * np.random.normal(size=x_train.shape) 
x_test_noisy = x_test + noise_factor * np.random.normal(size=x_test.shape)

x_train_noisy = np.clip(x_train_noisy, 0.0, 1.0)
x_test_noisy = np.clip(x_test_noisy, 0.0, 1.0)
print (x_train.shape)
print (x_test.shape)

# train our autoencoder for 50 epochs
print ("Training ....")
autoencoder.fit(x_train_noisy, x_train,epochs=10,batch_size=256,
                shuffle=True,
               )


# save output
path3 =  r'D:\result.csv'

output = autoencoder.predict(x_test_noisy)
with open(path3, 'w',  newline='') as f3, open(path2, 'r') as f4: # write output and otherr column from test
    forest_Csv = csv.writer(f3)
    forest_Csv.writerow(["Index","MachineID","Running","Started","Evict","Finish","Kill","Lost","Cpu","Memory","Disk time","CPI","MAI","CONCLUS","fault detection"])    
    test_file_object = csv.reader(f4)
    next(test_file_object, None)
    i = 0
    for row in  test_file_object:
        row.insert(15,output[i].astype(np.uint8))
        forest_Csv.writerow(row)
        i += 1

结果：返回一个12位数的序列。我不知道我哪里做错了，我不知道我是否正确。谁能帮我？非常感谢你 link train.csv和test.csv： https://drive.google.com/open?id=1DtXr5RnF8I3FDGN5DnIboO50K_iNughU

如何利用数据集csv进行深度稀疏去噪自动编码器的故障检测

0 个答案: