ValueError：logits 和标签必须具有相同的形状 ((None, 5) vs (None, 1))

Question

我正在为表格数据集使用一维 CNN，但是当我针对二元分类问题训练自定义模型时，出现错误。

下面是我的代码：

#splitting the datasets for training and testing process
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size =0.3, random_state=42)
#size for the sets
print('size of X_train:', X_train.shape)
print('size of X_test:', X_test.shape)
print('size of y_train:', y_train.shape)
print('size of y_test:', y_test.shape)

#CNN
from tensorflow.keras.layers import Flatten, Dense, Conv1D, MaxPool1D, Dropout
X_train = np.array(X_train).reshape(X_train.shape[0], X_train.shape[1], 1)
X_test = np.array(X_test).reshape(X_test.shape[0], X_test.shape[1], 1)
print("X Train shape: ", X_train.shape)
print("X Test shape: ", X_test.shape)

# Create sequential model 
cnn_model = tf.keras.models.Sequential()
#First CNN layer  with 32 filters, conv window 3, relu activation and same padding
cnn_model.add(Conv1D(filters=32, kernel_size=(3,), padding='same', activation=tf.keras.layers.LeakyReLU(alpha=0.001), input_shape = (X_train.shape[1],1)))
#Second CNN layer  with 64 filters, conv window 3, relu activation and same padding
cnn_model.add(Conv1D(filters=64, kernel_size=(3,), padding='same', activation=tf.keras.layers.LeakyReLU(alpha=0.001)))
#Third CNN layer with 128 filters, conv window 3, relu activation and same padding
cnn_model.add(Conv1D(filters=128, kernel_size=(3,), padding='same', activation=tf.keras.layers.LeakyReLU(alpha=0.001)))
#Fourth CNN layer with Max pooling
cnn_model.add(MaxPool1D(pool_size=(3,), strides=2, padding='same'))
cnn_model.add(Dropout(0.5))
#Flatten the output
cnn_model.add(Flatten())
#Add a dense layer with 256 neurons
cnn_model.add(Dense(units = 256, activation=tf.keras.layers.LeakyReLU(alpha=0.001)))
#Add a dense layer with 512 neurons
cnn_model.add(Dense(units = 512, activation=tf.keras.layers.LeakyReLU(alpha=0.001)))
#Softmax as last layer with five outputs
cnn_model.add(Dense(units = 5, activation='sigmoid'))
cnn_model.compile(optimizer='adam', loss = 'binary_crossentropy', metrics=['accuracy'])
cnn_model.summary()

cnn_model_history = cnn_model.fit(X_train, y_train, epochs=150, batch_size = 10, validation_data = (X_test, y_test))

我收到以下 logits 错误，并且标签必须具有相同的形状：

ValueError                                Traceback (most recent call last)
/tmp/ipykernel_34/2425696282.py in <module>
----> 1 cnn_model_history = cnn_model.fit(X_train, y_train, epochs=150, batch_size = 10, validation_data = (X_test, y_test))

/opt/conda/lib/python3.7/site-packages/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq, max_queue_size, workers, use_multiprocessing)
   1182                 _r=1):
   1183               callbacks.on_train_batch_begin(step)
-> 1184               tmp_logs = self.train_function(iterator)
   1185               if data_handler.should_sync:
   1186                 context.async_wait()

/opt/conda/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py in __call__(self, *args, **kwds)
    883 
    884       with OptionalXlaContext(self._jit_compile):
--> 885         result = self._call(*args, **kwds)
    886 
    887       new_tracing_count = self.experimental_get_tracing_count()

/opt/conda/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py in _call(self, *args, **kwds)
    931       # This is the first call of __call__, so we have to initialize.
    932       initializers = []
--> 933       self._initialize(args, kwds, add_initializers_to=initializers)
    934     finally:
    935       # At this point we know that the initialization is complete (or less

/opt/conda/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py in _initialize(self, args, kwds, add_initializers_to)
    758     self._concrete_stateful_fn = (
    759         self._stateful_fn._get_concrete_function_internal_garbage_collected(  # pylint: disable=protected-access
--> 760             *args, **kwds))
    761 
    762     def invalid_creator_scope(*unused_args, **unused_kwds):

/opt/conda/lib/python3.7/site-packages/tensorflow/python/eager/function.py in _get_concrete_function_internal_garbage_collected(self, *args, **kwargs)
   3064       args, kwargs = None, None
   3065     with self._lock:
-> 3066       graph_function, _ = self._maybe_define_function(args, kwargs)
   3067     return graph_function
   3068 

/opt/conda/lib/python3.7/site-packages/tensorflow/python/eager/function.py in _maybe_define_function(self, args, kwargs)
   3461 
   3462           self._function_cache.missed.add(call_context_key)
-> 3463           graph_function = self._create_graph_function(args, kwargs)
   3464           self._function_cache.primary[cache_key] = graph_function
   3465 

/opt/conda/lib/python3.7/site-packages/tensorflow/python/eager/function.py in _create_graph_function(self, args, kwargs, override_flat_arg_shapes)
   3306             arg_names=arg_names,
   3307             override_flat_arg_shapes=override_flat_arg_shapes,
-> 3308             capture_by_value=self._capture_by_value),
   3309         self._function_attributes,
   3310         function_spec=self.function_spec,

/opt/conda/lib/python3.7/site-packages/tensorflow/python/framework/func_graph.py in func_graph_from_py_func(name, python_func, args, kwargs, signature, func_graph, autograph, autograph_options, add_control_dependencies, arg_names, op_return_value, collections, capture_by_value, override_flat_arg_shapes, acd_record_initial_resource_uses)
   1005         _, original_func = tf_decorator.unwrap(python_func)
   1006 
-> 1007       func_outputs = python_func(*func_args, **func_kwargs)
   1008 
   1009       # invariant: `func_outputs` contains only Tensors, CompositeTensors,

/opt/conda/lib/python3.7/site-packages/tensorflow/python/eager/def_function.py in wrapped_fn(*args, **kwds)
    666         # the function a weak reference to itself to avoid a reference cycle.
    667         with OptionalXlaContext(compile_with_xla):
--> 668           out = weak_wrapped_fn().__wrapped__(*args, **kwds)
    669         return out
    670 

/opt/conda/lib/python3.7/site-packages/tensorflow/python/framework/func_graph.py in wrapper(*args, **kwargs)
    992           except Exception as e:  # pylint:disable=broad-except
    993             if hasattr(e, "ag_error_metadata"):
--> 994               raise e.ag_error_metadata.to_exception(e)
    995             else:
    996               raise

ValueError: in user code:

    /opt/conda/lib/python3.7/site-packages/keras/engine/training.py:853 train_function  *
        return step_function(self, iterator)
    /opt/conda/lib/python3.7/site-packages/keras/engine/training.py:842 step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/distribute/distribute_lib.py:1286 run
        return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/distribute/distribute_lib.py:2849 call_for_each_replica
        return self._call_for_each_replica(fn, args, kwargs)
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/distribute/distribute_lib.py:3632 _call_for_each_replica
        return fn(*args, **kwargs)
    /opt/conda/lib/python3.7/site-packages/keras/engine/training.py:835 run_step  **
        outputs = model.train_step(data)
    /opt/conda/lib/python3.7/site-packages/keras/engine/training.py:789 train_step
        y, y_pred, sample_weight, regularization_losses=self.losses)
    /opt/conda/lib/python3.7/site-packages/keras/engine/compile_utils.py:201 __call__
        loss_value = loss_obj(y_t, y_p, sample_weight=sw)
    /opt/conda/lib/python3.7/site-packages/keras/losses.py:141 __call__
        losses = call_fn(y_true, y_pred)
    /opt/conda/lib/python3.7/site-packages/keras/losses.py:245 call  **
        return ag_fn(y_true, y_pred, **self._fn_kwargs)
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py:206 wrapper
        return target(*args, **kwargs)
    /opt/conda/lib/python3.7/site-packages/keras/losses.py:1809 binary_crossentropy
        backend.binary_crossentropy(y_true, y_pred, from_logits=from_logits),
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py:206 wrapper
        return target(*args, **kwargs)
    /opt/conda/lib/python3.7/site-packages/keras/backend.py:5000 binary_crossentropy
        return tf.nn.sigmoid_cross_entropy_with_logits(labels=target, logits=output)
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py:206 wrapper
        return target(*args, **kwargs)
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/ops/nn_impl.py:246 sigmoid_cross_entropy_with_logits_v2
        logits=logits, labels=labels, name=name)
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/util/dispatch.py:206 wrapper
        return target(*args, **kwargs)
    /opt/conda/lib/python3.7/site-packages/tensorflow/python/ops/nn_impl.py:133 sigmoid_cross_entropy_with_logits
        (logits.get_shape(), labels.get_shape()))

    ValueError: logits and labels must have the same shape ((None, 5) vs (None, 1))
***

我的代码有什么问题，有人有解决方案吗？

Answer 1

如果您的标签是稀疏整数，其中每个样本属于 {0, 1, 2, 3, 4} 五个类之一，那么您应该在输出层和 SparseCategoricalCrossentropy() 上使用 softmax 激活函数：

cnn_model.add(Dense(units = 5, activation='softmax'))
cnn_model.compile(optimizer='adam', loss = tf.keras.losses.SparseCategoricalCrossentropy(), metrics=['accuracy'])

如果您正在处理每个样本可以属于 0 类或 1 类的二元分类问题，那么您需要将最后一层更改为只有一个输出节点：

cnn_model.add(Dense(units = 1, activation='sigmoid'))

如果您正在处理一个多标签分类问题，其中每个标签可以属于多个标签，例如[1, 1, 0, 1, 0]，那么您需要对您的标签进行一次热编码，以便每个标签由 5 个表示存在的二进制数组成或每个样本的每个类别不存在。您的模型本身可以保持不变。