Python模型训练出现ValueError：logits与labels形状不匹配

错误核心原因 #

你的模型输出层维度为10（对应(None,10)的logits），但训练标签维度是1（(None,1)），同时使用了binary_crossentropy损失函数，三者不匹配导致报错。

针对性解决方案 #

场景1：实际做多分类任务（10个类别） #

如果任务是区分10个不同类别，调整损失函数和标签格式即可：

• 若标签是独热编码格式（每个样本标签为长度10的向量，仅对应类别位为1）：
  • 模型编译时使用categorical_crossentropy损失函数
  • 模型最后一层保持Dense(10, activation='softmax')
• 若标签是整数索引格式（每个样本标签为0-9的单个整数，对应(None,1)形状）：
  • 模型编译时使用sparse_categorical_crossentropy损失函数
  • 模型最后一层保持Dense(10, activation='softmax')

示例代码：

# 模型最后一层（多分类输出）
model.add(Dense(10, activation='softmax'))

# 标签为整数索引时编译
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])

# 标签为独热编码时编译
model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])

场景2：实际做二分类任务 #

如果任务是区分两个类别，需要调整模型输出层：

• 模型最后一层改为输出维度1，激活函数用sigmoid
• 保持binary_crossentropy损失函数，标签(None,1)的形状无需修改

示例代码：

# 修改模型最后一层为二分类输出
model.add(Dense(1, activation='sigmoid'))

# 编译模型（损失函数保持binary_crossentropy）
model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])

错误日志参考 #

第1轮训练/共100轮

ValueError Traceback (most recent call last)
在<cell line: 1>()中
----> 1 model_history=classifier.fit(X_train,Y_train,batch_size=100,validation_split=0.2,epochs = 100)

1 frames
/usr/local/lib/python3.9/dist-packages/keras/engine/training.py in tf__train_function(iterator)
13 try:
14 do_return = True
---> 15 retval_ = ag__.converted_call(ag__.ld(step_function), (ag__.ld(self), ag__.ld(iterator)), None, fscope)
16 except:
17 do_return = False

ValueError: 用户代码中出现错误：

File "/usr/local/lib/python3.9/dist-packages/keras/engine/training.py", line 1284, in train_function  *
    return step_function(self, iterator)
File "/usr/local/lib/python3.9/dist-packages/keras/engine/training.py", line 1268, in step_function  **
    outputs = model.distribute_strategy.run(run_step, args=(data,))
File "/usr/local/lib/python3.9/dist-packages/keras/engine/training.py", line 1249, in run_step  **
    outputs = model.train_step(data)
File "/usr/local/lib/python3.9/dist-packages/keras/engine/training.py", line 1051, in train_step
    loss = self.compute_loss(x, y, y_pred, sample_weight)
File "/usr/local/lib/python3.9/dist-packages/keras/engine/training.py", line 1109, in compute_loss
    return self.compiled_loss(
File "/usr/local/lib/python3.9/dist-packages/keras/engine/compile_utils.py", line 265, in __call__
    loss_value = loss_obj(y_t, y_p, sample_weight=sw)
File "/usr/local/lib/python3.9/dist-packages/keras/losses.py", line 142, in __call__
    losses = call_fn(y_true, y_pred)
File "/usr/local/lib/python3.9/dist-packages/keras/losses.py", line 268, in call  **
    return ag_fn(y_true, y_pred, **self._fn_kwargs)
File "/usr/local/lib/python3.9/dist-packages/keras/losses.py", line 2156, in binary_crossentropy
    backend.binary_crossentropy(y_true, y_pred, from_logits=from_logits),
File "/usr/local/lib/python3.9/dist-packages/keras/backend.py", line 5707, in binary_crossentropy
    return tf.nn.sigmoid_cross_entropy_with_logits(

ValueError: `logits` and `labels` must have the same shape, received ((None, 10) vs (None, 1)).

ValueError: logits and labels must have the same shape, received ((None, 10) vs (None, 1)).

内容的提问来源于stack exchange，提问作者Balasubramanian Vellaichamy