输入为(250,5000),其中每个示例为 250 个样本,我们有 5000 个示例。
import scipy.io
X = scipy.io.loadmat("/drive/My Drive/Cola Notebooks/x_train.mat")
Y = scipy.io.loadmat("/drive/My Drive/Colab Notebooks/y_train.mat")
X_train = X['x_train']
Y_train = Y['y_train']
model = keras.Sequential([
keras.layers.Flatten(input_shape=(250, 1)),
keras.layers.Dense(128, activation='relu'),
keras.layers.Dense(10, activation='softmax')
])
model.compile(optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'])
model.fit(X_train, Y_train, epochs=10)
编译模型后。适合,我收到以下错误ValueError:
检查输入时出错:预期
的数组flatten_10_input
为3 尺寸,但形状为(250,5000)