I am trying to use this to build my own fully connected layer.

But I meet a problem here is that how could I retrieve the output of the previous layer? In keras, there is a flatten layer followed the convolution layer.

And in this case, how could Matlab recognize the shape of the output layer?

Could you please help me? ]]>

Thanks and Regards,

Sunny

Avi

]]>This is a good introduction to object detection. Over the past year I have been exploring a lot of MATLAB tutorials and find it excellent. Keep up the great content.

I was wondering if MATLAB had support for other DNN frameworks like exporting a Caffe model or support for YOLO ? It would be great, if a tool exists where I can use these approaches in MATLAB. The reason I ask is because my deployment target is defined and cannot be changed.

Thanks.

]]>fs = 300;]]>