soonmok / mask-rcnn-tf-2.0 Goto Github PK

While using this repository for custom dataset, I encountered the below error on training the model:

ValueError: Duplicate node name in graph: 'mask_rcnn/proposal_targets/packed'

TF version: 2.0
Keras 2.3.1
Python 3.6

I want to ask a silly question as to how to use this tool to fix the masked rcnn? I do not see setup.py in it. Thanks

Hi,

Thanks for your good work.

I came across your repository and thought of replacing existing MRCNN (by MatterPort) with yours. So I firked it and made some changes to make it a drop-in replacement. I am stuck with the following error: Now the line number might be slightly different from the one in your code but not too much.

I thought of creating an issue here while I try to find solution (I have little experience in TF 2.0).

Thanks

File "/code/third-party/mask-rcnn-tf-2.0/mrcnn/model.py", line 570, in train
self.compile(learning_rate, self.config.LEARNING_MOMENTUM)
File "/code/third-party/mask-rcnn-tf-2.0/mrcnn/model.py", line 388, in compile
if tf.reduce_any(input_tensor=[layer.output is loss for loss in self.keras_model.losses]):
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/framework/ops.py", line 765, in bool
self._disallow_bool_casting()
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/framework/ops.py", line 534, in _disallow_bool_casting
self._disallow_in_graph_mode("using a tf.Tensor as a Python bool")
File "/usr/local/lib/python3.6/dist-packages/tensorflow_core/python/framework/ops.py", line 523, in _disallow_in_graph_mode
" this function with @tf.function.".format(task))
tensorflow.python.framework.errors_impl.OperatorNotAllowedInGraphError: using a tf.Tensor as a Python bool is not allowed in Graph execution. Use Eager execution or decorate this function with @tf.function.