keras模型训练输出和测试输出数据维度不一致

Question

使用keras编写了一个神经网络做预测，训练时输入数据维度是(1736, 30, 1)，输出数据维度是(1736, 1)。
但是模型训练完成后测试输入数据维度是(60, 30, 1)，此时模型输出数据的维度变成了(60, 30, 1)。
输出从二维变成了三维，请问这是为什么，需要怎么解决啊。

Xtrain = X[:k, :, :]    # (1736, 30, 1)
Xtest = X[k:, :, :]     # (60, 30, 1)
Ytrain = Y[:k]          # (1736, 1)
Ytest = Y[k:]           # (60, 1)
print(X.shape, Y.shape, Xtrain.shape, Xtest.shape, Ytrain.shape, Ytest.shape)
model = Sequential()
model.add(Conv1D(filters=256, kernel_size=2, activation='relu', input_shape=(30, 1)))
model.add(Conv1D(filters=128, kernel_size=2, activation='relu'))
model.add(MaxPooling1D(pool_size=2))
model.add(Flatten())
model.add(RepeatVector(30))
model.add(LSTM(units=100, return_sequences=True))
model.add(Dropout(0.2))
model.add(LSTM(units=100, return_sequences=True))
model.add(LSTM(units=100, return_sequences=True))
# model.add(Bidirectional(LSTM(128, activation='relu')))
model.add(Dense(100, activation='relu'))
model.add(Dense(1))
model.compile(loss='mse', optimizer='adam')
history = model.fit(Xtrain, Ytrain, epochs=10, verbose=1)

# model.save("./regressor.hdf5")

# model = load_model('./regressor.hdf5')
predict = model.predict(Xtest)
print(predict.shape)        # (60, 30, 1)

Answer 1

试下这样：
predict2 = model.predict(Xtrain)
print(predict2.shape)
看下是否训练数据也一样