我定义了模型如下。
tf.keras.datasets.mnist model = keras.models.Sequential([ tf.keras.layers.Conv2D(28, (3,3), activation='relu', input_shape=(28, 28, 1)), tf.keras.layers.MaxPooling2D((2, 2)), tf.keras.layers.Conv2D(56, (3,3), activation='relu'), tf.keras.layers.Flatten(), tf.keras.layers.Dense(64, activation='relu'), tf.keras.layers.Dense(10, activation='softmax'), ]) model.fit(x_train, y_train, epochs=3) <----错误当我尝试使用我的数据集运行时，出现了以下错误。
ValueError: 层 sequential_3 的输入 0 与层不兼容：预期的最小维度为4，但发现维度为3。接收到的完整形状为：[32, 28, 28]回答：
从您的错误来看，我认为您可能没有在训练集中添加通道轴，即 [batch, w, h, channel]。这是工作代码
数据集
import tensorflow as tf import numpy as np from sklearn.model_selection import train_test_split(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()x_train = np.expand_dims(x_train, axis=-1) # <--- 添加通道轴x_train = x_train.astype('float32') / 255y_train = tf.keras.utils.to_categorical(y_train, num_classes=10)print(x_train.shape, y_train.shape)# (60000, 28, 28, 1) (60000, 10)训练
model = tf.keras.models.Sequential([ tf.keras.layers.Conv2D(28, (3,3), activation='relu', input_shape=(28, 28, 1)), tf.keras.layers.MaxPooling2D((2, 2)), tf.keras.layers.Conv2D(56, (3,3), activation='relu'), tf.keras.layers.Flatten(), tf.keras.layers.Dense(64, activation='relu'), tf.keras.layers.Dense(10, activation='softmax'), ]) model.compile(loss='categorical_crossentropy',optimizer='Adam',metrics=['accuracy'])model.fit(x_train, y_train, epochs=3)第1轮/共3轮1875/1875 [==============================] - 11s 2ms/step - 损失: 0.2803 - 准确率: 0.9160第2轮/共3轮1875/1875 [==============================] - 4s 2ms/step - 损失: 0.0434 - 准确率: 0.9869第3轮/共3轮1875/1875 [==============================] - 4s 2ms/step - 损失: 0.0273 - 准确率: 0.9917