Concat Layer Fold Constant Error

[junhui-ng]

I have a CONCAT layer with four inputs — three of which are constants, and one is the output from a TRANSPOSE layer. While converting the model from ONNX to RKNN using the Python API, an error occurred:

E build: Traceback (most recent call last): 
  File "rknn/api/rknn_log.py", line 344, in rknn.api.rknn_log.error_catch_decorator.error_catch_wrapper  
  File "rknn/api/rknn_base.py", line 2007, in rknn.api.rknn_base.RKNNBase.build 
  File "rknn/api/graph_optimizer.py", line 2338, in rknn.api.graph_optimizer.GraphOptimizer.fuse_ops 
  File "rknn/api/graph_optimizer.py", line 960, in rknn.api.graph_optimizer.GraphOptimizer.fold_constant 
KeyError: '/decoder/ifi_1/Transpose_4_output_0'

The red line indicates where the error occurred:

And this is the information of the CONCAT layer: