• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 100 subscribers
  • Views 130 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

[FAQ] SK-TDA4VM: How to do inferencing with custom compiled model on target

SK-TDA4VM: how to enable single dataset in ONR-OD-8200-yolox-nano-lite-mmdet-coco-416x416 model of SK-TDA4VM

Chaithra B V
Prodigy 80 points
Part Number: SK-TDA4VM

Hello,

I am working on enabling the single data-set (person) in the "ONR-OD-8200-yolox-nano-lite-mmdet-coco-416x416" model for object detection 

for that, I have keep one dataset and removed all other data-set and I run my object detection. 

Thowing a below error:-

terminate called after throwing an instance of '[ 6157.158406] audit: type=1701 audit(1698908663.080:25): auid=4294967295 uid=0 gid=0 s
es=4294967295 pid=2350 comm="app_edgeai" exe="/opt/edgeai-gst-apps/apps_cpp/bin/Release/app_edgeai" sig=6 res=1
std::out_of_range'
                    what():  map::at
please, help me out to enable the single dataset?
  • 0
    TI__Mastermind 24041 points

    Hi,

    If I understand your question correctly, you want to detect only single object in an image, from yolox model trained on multiclass dataset ?

    Am not completely sure on one can modify the dataset.yaml file to delete other classes indexes, I hardly think this is advisable  as these file are have nested dependencies wrt edgeai gst apps flow.

    You can use the model file which has single detection prediction and trained for detection of specific object, we recommend to use tidl tools offering to compile the model and do the inference.