Hi all,
I have above mentioned two meraki cameras installed in my office.
I want to test custom yolov8 models on these two cameras but facing some issues with that.
Following are the issues:
1. Input image should be normalized between -1 to 1
1. The output of yolov8 is in different format i.e [ 1,84,8400] but Meraki expects "Output" section format in below document
https://documentation.meraki.com/MV/Video_Analytics/MV_Sense_Custom_Computer_Vision
Now, is there any way I can counter these issues either from
1. Pre-processing and post-processing using custom code within miraki dashboard or any other way
2. Changing the model architecture itself by adding new output layer (if yes, can you please guide me through this)
