拟合 RNN LSTM 模型时出错

Question

我正在尝试使用以下代码为二进制分类创建 RNN LSTM 模型

alldataset = np.loadtxt("FinalKNEEALL.txt", delimiter=",")
num_classes = 2
num_of_sam = alldataset.shape[0]
labels = np.ones((num_of_sam,), dtype='int64')
labels[0:958943]=0
labels[958943:1917887]=1
Y = np_utils.to_categorical(labels,num_classes)
x,y = shuffle (alldataset,Y, random_state=2)
x_train,x_test, y_train,y_test = train_test_split(x,y, test_size=0.3, random_state=4)

print(x_train.shape)
print(x_test.shape)
print(y_train.shape)
print(y_test.shape)

x_train = x_train[:,[1,2,3,4,5,6]]
x_test = x_test[:,[1,2,3,4,5,6]]
y_train = y_train[:,0]
y_test = y_test[:,0]

print(x_train.shape)
print(x_test.shape)
print(y_train.shape)
print(y_test.shape)
input_width = 32
def windowz(data, size):
    start = 0
    while start < len(data):
        yield start, start + size
        start += (size // 2)

def segment_dap(x_train,y_train,window_size):
    segments = np.zeros(((len(x_train)//(window_size//2))-1,window_size,6))
    labels = np.zeros(((len(y_train)//(window_size//2))-1))
    i_segment = 0
    i_label = 0
    for (start,end) in windowz(x_train,window_size):
        if(len(x_train[start:end]) == window_size):
            m = stats.mode(y_train[start:end])
            segments[i_segment] = x_train[start:end]
            labels[i_label] = m[0]
            i_label+=1
            i_segment+=1
    return segments, labels
train_x, train_y = segment_dap(x_train,y_train,input_width)
test_x, test_y = segment_dap(x_test,y_test,input_width)

print(train_x.shape)
print(train_y.shape)
print(test_x.shape)
print(test_y.shape)

model = Sequential()
model.add(LSTM(64, input_shape=(32, 6), kernel_initializer = 'normal',
               activation='tanh'))
model.add(Dense(32, kernel_initializer = 'normal', activation='sigmoid' ))
model.compile(loss='binary_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])


hist = model.fit(train_x, train_y, batch_size = 30, epochs = 15,
                 verbose = 2
                 #validation_data=(test_x, test_y)
                 )

但它给了我以下错误

ValueError: Error when checking target: expected dense_1 to have shape (32,) but got array with shape (1,)

在尝试解决错误时，我将“binary_crossentropy”替换为“sparse_categorical_crossentropy”，但它使准确度达到 63%。此错误是否有其他解决方案？

这是我的数据集的示例

#(patient number, time in ms, normalization of X Y and Z,kurtosis, skewness, 
#pitch, roll and yaw, label(0 or 1)) respectively.

1,15,-0.248010047716,0.00378335508419,-0.0152548459993,-86.3738760481,0.872322164158,-3.51314800063,0
1,31,-0.248010047716,0.00378335508419,-0.0152548459993,-86.3738760481,0.872322164158,-3.51314800063,0
1,46,-0.267422664673,0.0051143782875,-0.0191247001961,-85.7662354031,1.0928406847,-4.08015176908,0
1,62,-0.267422664673,0.0051143782875,-0.0191247001961,-85.7662354031,1.0928406847,-4.08015176908,0

这是我从代码中打印出来的数据集的形状

(1342520, 8)
(575367, 8)
(1342520, 2)
(575367, 2)
(1342520, 6)
(575367, 6)
(1342520,)
(575367,)
(83906, 32, 6)
(83906,)
(35959, 32, 6)
(35959,)

Answer 1

模型的最后一层有 32 个单位Dense(32,... )。假设您正在进行二进制分类，最后一层应该有 1 个具有 sigmoid 激活的单元。

model = Sequential()
model.add(LSTM(64, input_shape=(32, 6), kernel_initializer = 'normal',
               activation='tanh'))
model.add(Dense(32, kernel_initializer = 'normal', activation='relu' ))
model.add(Dense(1,kernel_initializer = 'normal', activation="sigmoid")
model.compile(loss='binary_crossentropy',
              optimizer='adam',
              metrics=['accuracy'])