mxnet.gluon.parameter.DeferredInitializationError: Parameter 'XXX' has not been initialized yet.

• 错误信息：

mxnet.gluon.parameter.DeferredInitializationError: Parameter 'vggnet_fpn0_conv19_weight' has not been initialized yet because initializat
ion was deferred. Actual initialization happens during the first forward pass. Please pass one batch of data through the network before a
ccessing Parameters. You can also avoid deferred initialization by specifying in_units, num_features, etc., for network layers.

• 解决方案
• 检查Class的init中是否存在未在forward中使用的层的定义。
• 对conv等层传递参数时，确认参数传递正确，可以加上参数名来进行传递，e.g.,

self.c3_out.add(Conv2dTranspose(channels=256,
                                kernel_size=4,
                                strides=4,
                                in_channels=64))