python - Keras:方法 on_batch_end() 很慢但我没有回调？

Question

Keras 刚刚警告我:

UserWarning: Method on_batch_end() is slow compared to the batch update (0.988304). Check your callbacks.

我的 keras 脚本中没有回调，但我正在使用 DataGenerator (keras.utils.Sequence)。这会是个问题吗？

对于DataGenerator，我实现了__init__()、__len__()、__getitem()__和 on_epoch_end 方法。

对于 on_epoch_end，我有:

def on_epoch_end(self):
    """
    This method will be called between every epoch, so we can shuffle the
    indexes here.
    """
    self.indexes = np.arange(len(self.image_names))
    if self.shuffle:
        np.random.shuffle(self.indexes)

完整的调用栈如下:

Using TensorFlow backend. /var/lib/condor/execute/slot1/dir_30551/anaconda/envs/cellimage/lib/python3.6/site-packages/skimage/transform/_warps.py:84: UserWarning: The default mode, 'constant', will be changed to 'reflect' in skimage 0.15.

warn("The default mode, 'constant', will be changed to 'reflect' in " /var/lib/condor/execute/slot1/dir_30551/anaconda/envs/cellimage/lib/python3.6/site-packages/keras/callbacks.py:120: UserWarning: Method on_batch_end() is slow compared to the batch update (0.586719). Check your callbacks. % delta_t_median)

/var/lib/condor/execute/slot1/dir_30551/anaconda/envs/cellimage/lib/python3.6/site-packages/keras/callbacks.py:120: UserWarning: Method on_batch_end() is slow compared to the batch update (0.988304). Check your callbacks. % delta_t_median)

Answer 1

这可能与您的 verbose 设置有关。

我看到的是一样的。我相信这是因为我在调用 fit() 时有 verbose=1 - 即打印进度条和当前统计信息是一个批处理结束任务，它也需要长的。如果我设置仅在纪元结束时打印的 verbose=2 或根本不打印的 verbose=0 ，警告就会停止，这一假设得到了证据的支持。如果我增加批量大小，警告也会停止。