當我定義一個函數來執行神經網絡訓練時，我得到一個錯誤。但是，當我這樣做而不使它起作用時，就沒有錯誤。為什么？def myneural():    import numpy as np    import keras    from keras import backend as K    from keras.layers import Input, Dense, Activation    from keras.models import Model, Sequential    x_train_s = np.random.randint(5, size=(20, 2))    x_test_s = x_train_s    model = Sequential([        Dense(10, input_shape=(2,)),        Activation('linear'),        Dense(2),        Activation('linear')    ])    model.compile(optimizer='adam', loss='mean_squared_error')    fittingadam = model.fit(x_train_s, x_train_s, epochs=2,                            validation_data=(x_test_s, x_test_s),                            shuffle=True, verbose=1, batch_size=None)    encoder = K.function([model.layers[0].input],                         [model.layers[1].output])    code = encoder([x_test_s])[0]myneural()我得到的錯誤是：Using TensorFlow backend.WARNING:tensorflow:From C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\framework\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.Instructions for updating:Colocations handled automatically by placer.WARNING:tensorflow:From C:\ProgramData\Anaconda3\lib\site-packages\tensorflow\python\ops\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.Instructions for updating:Use tf.cast instead.Train on 20 samples, validate on 20 samplesEpoch 1/22019-10-03 14:34:50.275279: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX220/20 [==============================] - 0s 7ms/step - loss: 4.0432 - val_loss: 3.9670Epoch 2/2但是，當我刪除第一行和最后一行并從其他行中刪除縮進時，不會有錯誤。第一個問題：為什么會這樣？我怎樣才能解決它以運行它而沒有任何問題作為函數？第二個問題：警告呢？它們重要嗎？我怎樣才能擺脫它們？