我对机器学习和Keras都还比较新手。我试图创建一个模型，该模型可以接受一个数组的数组列表作为输入（一个包含6400个数组的列表，每个数组内有2个数组）。这是我代码的问题：

XFIT = np.array([x_train, XX_train])YFIT = np.array([y_train, yy_train])Inputs = keras.layers.Input(shape=(6400, 2))hidden1 = keras.layers.Dense(units=100, activation="sigmoid")(Inputs)hidden2 = keras.layers.Dense(units=100, activation='relu')(hidden1)predictions = keras.layers.Dense(units=3, activation='softmax')(hidden2)model = keras.Model(inputs=Inputs, outputs=predictions)

没有错误；然而，输入层（Inputs）强制我传递一个形状为（6400, 2）的输入，因为每个数组（x_train和XX_train）内部都有6400个数组。经过若干轮次后的结果是这样的：

Train on 2 samplesEpoch 1/52/2 [==============================] - 1s 353ms/sample - loss: 1.1966 - accuracy: 0.2488Epoch 2/52/2 [==============================] - 0s 9ms/sample - loss: 1.1303 - accuracy: 0.2544Epoch 3/52/2 [==============================] - 0s 9ms/sample - loss: 1.0982 - accuracy: 0.3745Epoch 4/52/2 [==============================] - 0s 9ms/sample - loss: 1.0854 - accuracy: 0.3745Epoch 5/52/2 [==============================] - 0s 9ms/sample - loss: 1.0835 - accuracy: 0.3745Process finished with exit code 0

由于输入形状的原因，我无法在每个轮次中训练超过两次。我该如何改变这个输入？我尝试过其他形状，但都得到了错误。

x_train, XX_train看起来像这样

[[[0.505834 0.795461]  [0.843175 0.975741]  [0.22349  0.035036]  ...  [0.884796 0.867509]  [0.396942 0.659936]  [0.873194 0.05454 ]] [[0.95968  0.281957]  [0.137547 0.390005]  [0.635382 0.901555]  ...  [0.887062 0.486206]  [0.49827  0.949123]  [0.034411 0.983711]]]

谢谢你，如果我犯了任何错误，请原谅我，这是我在Keras和StackOverflow的第一次尝试 😀

回答：

你已经接近成功了。问题在于：

XFIT = np.array([x_train, XX_train])YFIT = np.array([y_train, yy_train])

让我们通过一个例子来看：

import numpy as npx_train = np.random.random((6400, 2))y_train = np.random.randint(2, size=(6400,1))xx_train = np.array([x_train, x_train])yy_train = np.array([y_train, y_train])print(xx_train.shape)(2, 6400, 2)print(yy_train.shape)(2, 6400, 1)

在这个数组中，我们有2个批次，每个批次包含6400个样本。这意味着当我们调用model.fit时，它只有2个批次可以训练。相反，我们可以这样做：

xx_train = np.vstack([x_train, x_train])yy_train = np.vstack([y_train, y_train])print(xx_train.shape)(12800, 2)print(yy_train.shape)(12800, 1)

现在，我们已经正确地合并了两个样本，现在可以进行训练了。

Inputs = Input(shape=(2, ))hidden1 = Dense(units=100, activation="sigmoid")(Inputs)hidden2 = Dense(units=100, activation='relu')(hidden1)predictions = Dense(units=1, activation='sigmoid')(hidden2)model = Model([Inputs], outputs=predictions)model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])model.fit(xx_train, yy_train, batch_size=10, epochs=5)Train on 12800 samplesEpoch 1/512800/12800 [==============================] - 3s 216us/sample - loss: 0.6978 - acc: 0.5047Epoch 2/512800/12800 [==============================] - 2s 186us/sample - loss: 0.6952 - acc: 0.5018Epoch 3/512800/12800 [==============================] - 3s 196us/sample - loss: 0.6942 - acc: 0.4962Epoch 4/512800/12800 [==============================] - 3s 217us/sample - loss: 0.6938 - acc: 0.4898Epoch 5/512800/12800 [==============================] - 3s 217us/sample - loss: 0.6933 - acc: 0.5002