python - 如何插入 numpy 数组以适合我的模型？

Question

我已经创建了一个模型，input_shape(28,28,3)并且我有一numpy组数据需要适合shape(28500, 784).

谁能帮我将训练数据拟合到我的模型中？

model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(28,28,3)))

x_train.shape 

输出 = (28500, 784)

标签： pythonnumpytensorflowkerasartificial-intelligence

Answer 1

new_train = []
nchannels = 3
for x in x_train:
  x = np.resize(x, (28, 28, nchannels))
  new_train.append(x)
new_train = np.array(new_train)