tf.keras.layers.SpatialDropout3D(rate, data_format=None, **kwargs)
Spatial 3D version of Dropout.
This version performs the same function as Dropout, however, it drops entire 3D feature maps instead of individual elements. If adjacent voxels within feature maps are strongly correlated (as is normally the case in early convolution layers) then regular dropout will not regularize the activations and will otherwise just result in an effective learning rate decrease. In this case, SpatialDropout3D will help promote independence between feature maps and should be used instead.
image_data_formatvalue found in your Keras config file at
~/.keras/keras.json. If you never set it, then it will be "channels_last".
5D tensor with shape:
(samples, channels, dim1, dim2, dim3) if
or 5D tensor with shape:
(samples, dim1, dim2, dim3, channels) if
Output shape Same as input