我在Keras中训练了一个回归问题的神经网络.

like this:

1000/199873 [..............................] - 预计到达时间:5秒-损失:0.0057-附件:0.0000e+00

2000/199873 [..............................] - 预计到达时间:4s-损失:0.0058-附件:0.0000e+00

3000/199873 [..............................] - 预计到达时间:3s-损失:0.0057-附件:0.0000e+00

4000/199873 [..............................] - 预计到达时间:3秒-损失:0.0060-附件:

198000/199873 [============================>.] - ETA: 0s - loss: 0.0055 - acc: 0.0000e+00

199000/199873 [============================>.] - ETA: 0s - loss: 0.0055 - acc: 0.0000e+00

199873/199873[======================================================================================================-4s-损耗:0.0055-附件:0.0000e+00-损耗:0.0180-附件:0.0000e+00

Epoch 50/50

但如果输出是二维或更高的，则精度没有问题.

我的模型如下:`

input_dim = 14
batch_size = 1000
nb_epoch = 50
lrelu = LeakyReLU(alpha = 0.1)

model = Sequential()
model.add(Dense(126, input_dim=input_dim)) #Dense(output_dim(also hidden wight), input_dim = input_dim)
model.add(lrelu) #Activation

model.add(Dense(252))
model.add(lrelu)
model.add(Dense(1))
model.add(Activation('linear'))

model.compile(loss= 'mean_squared_error', optimizer='Adam', metrics=['accuracy'])
model.summary()
history = model.fit(X_train_1, y_train_1[:,0:1],
                    batch_size=batch_size,
                    nb_epoch=nb_epoch,
                    verbose=1,
                    validation_split=0.2)

loss = history.history.get('loss')
acc = history.history.get('acc')
val_loss = history.history.get('val_loss')
val_acc = history.history.get('val_acc')

'''saving model'''
from keras.models import load_model
model.save('XXXXX')
del model

'''loading model'''
model = load_model('XXXXX')

'''prediction'''
pred = model.predict(X_train_1, batch_size, verbose=1)
ans = [np.argmax(r) for r in y_train_1[:,0:1]]