预测经过训练的 DNN 模型的类别时出错

Question

我使用以下程序来预测图像的类别。from tensorflow.keras.preprocessing.image import load_img, img_to_arrayx = load_img("8-SignLanguageMNIST/test1.jpg", target_size = (28, 28))x = img_to_array(x)x = np.expand_dims(x, axis = 0)x = np.vstack([x])classes = model.predict(x)print(classes[0])我用于训练的图像形状为 (28, 28, 1)。在这里，我上传了一个形状为 (28, 28, 3) 的 RGB 图像，我尝试将该图像转换为灰度，然后进行预测，但一直出现以下错误。ValueError: Input 0 of layer sequential is incompatible with the layer: expected axis -1 of input shape to have value 1 but received input with shape [None, 28, 28, 3]谁能告诉我我做错了什么，并帮助我解决这个问题。

Answer 1

您需要将转换应用到灰度，如下所示：

load_img(path, color_mode='grayscale')