keras - 1D-CNN 输入兼容性错误，时间序列问题答案

【问题标题】：keras - 1D-CNN input compatibility error, a time series problemkeras - 1D-CNN 输入兼容性错误，时间序列问题
【发布时间】：2021-09-27 05:59:06
【问题描述】：

我正在使用下面的代码来训练一维 CNN，我的 x_train 数据形状是 (10027, 5, 14)，y_train 形状是 (10027,4)。但我收到一个错误（在代码下方）关于形状兼容性。


model = keras.models.Sequential([
    keras.layers.Conv1D(filters=20, kernel_size=4, strides=2, padding="valid",
                        input_shape=(n_timesteps,n_features))
  
])
model.compile(loss=keras.losses.categorical_crossentropy,
              optimizer=keras.optimizers.Adadelta(),
              metrics=['accuracy'])

 verbose, epochs, batch_size = 0, 10, 5 



 model.fit(X_train, Y_train, epochs=epochs, batch_size=batch_size, verbose=verbose)```

Below error 

```ValueError: in user code:

    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/engine/training.py:805 train_function  *
        return step_function(self, iterator)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/engine/training.py:795 step_function  **
        outputs = model.distribute_strategy.run(run_step, args=(data,))
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/distribute/distribute_lib.py:1259 run
        return self._extended.call_for_each_replica(fn, args=args, kwargs=kwargs)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/distribute/distribute_lib.py:2730 call_for_each_replica
        return self._call_for_each_replica(fn, args, kwargs)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/distribute/distribute_lib.py:3417 _call_for_each_replica
        return fn(*args, **kwargs)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/engine/training.py:788 run_step  **
        outputs = model.train_step(data)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/engine/training.py:755 train_step
        loss = self.compiled_loss(
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/engine/compile_utils.py:203 __call__
        loss_value = loss_obj(y_t, y_p, sample_weight=sw)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/losses.py:152 __call__
        losses = call_fn(y_true, y_pred)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/losses.py:256 call  **
        return ag_fn(y_true, y_pred, **self._fn_kwargs)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper
        return target(*args, **kwargs)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/losses.py:1537 categorical_crossentropy
        return K.categorical_crossentropy(y_true, y_pred, from_logits=from_logits)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/util/dispatch.py:201 wrapper
        return target(*args, **kwargs)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/keras/backend.py:4833 categorical_crossentropy
        target.shape.assert_is_compatible_with(output.shape)
    /usr/local/lib/python3.8/dist-packages/tensorflow/python/framework/tensor_shape.py:1134 assert_is_compatible_with
        raise ValueError("Shapes %s and %s are incompatible" % (self, other))

    ValueError: Shapes (None, 4) and (None, 1, 20) are incompatible```

【问题讨论】：

标签： python tensorflow keras time-series conv-neural-network

【解决方案1】：

我的意思是，你的网络的最后一层是一个 conv1D，filters=20，所以你的网络的输出将是 (batch, 1, 20)，你给出一个形状为 (batch, 4) 的 Y .

损失不明白它要比较什么，因为这两个数组根本不匹配。

如果你想让某些东西正常工作，你应该给最后一层与你的 Y 兼容：

model = keras.models.Sequential([
    keras.layers.Conv1D(filters=20, kernel_size=4, strides=2, padding="valid",
                        input_shape=(n_timesteps,n_features)),
    keras.layers.Flatten(),
    keras.layers.Dense(4, activation = 'softmax')  
])

【讨论】：