Warning below shows when i tray to load a trained RCNN Object Detector from a .mat file

Warning: During load:

An invalid default object has been detected while loading a heterogeneous array of class

nnet.cnn.layer.Layer. An empty array of class nnet.cnn.layer.Layer will be returned.

Warning: While loading an object of class 'DAGNetwork':

Array indices must be positive integers or logical values.

Warning: While loading an object of class 'fasterRCNNObjectDetector':

Dot indexing is not supported for variables of this type.

any advice to solve this problem

thanks in advince

Answer 1

[Yash]

I encountered the same issue while experimenting with the example Object Detection Using Faster R-CNN Deep Learning. After running the example in MATLAB R2020a, I successfully generated the .MAT file fasterRCNNResNet50EndToEndVehicleExample.mat and was able to load it without any problems in R2020a.

However, when attempting to load the same file in MATLAB R2018a, I received the identical error message. It appears that a Faster R-CNN model trained in R2020a is not compatible with R2018a.

To address this, one potential solution is to train the model directly in R2018a and subsequently load the .MAT file within the same version. Alternatively, upgrading to a later release of MATLAB and executing the entire process there may also resolve the issue.