Skip to content

Model requirement for vision ML demos on SL2619 #36

@prabrit-txr

Description

@prabrit-txr

I am using SL2610 series RDK having SL2619 inside.
I am following the steps mentioned in the git repo to run the image classification, object detection and pose estimation.
I am getting the following errors:

Image classification:
(.venv) root@sl2619:~/examples# python -m vision.image_class './samples/fish.jpg'
FileNotFoundError: [Errno 2] No such file or directory: '/usr/share/synap/models/image_classification/imagenet/info.json'

Object detection:
python -m vision.object_detect 'out.jpg'
[ ERROR ] synapRT.utils.input:L80: Error: SyNAP model '/usr/share/synap/models/object_detection/coco/model/yolov8s-640x384/model.synap' not found

Pose estimation
python -m vision.body_pose 'cam'
[ ERROR ] synapRT.utils.input:L80: Error: SyNAP model '/usr/share/synap/models/object_detection/body_pose/model/yolov8s-pose/model.synap' not found

Some more info:

(.venv) root@sl2619:~/examples# cat /etc/astra_version
scarthgap_6.12_v2.2.0

(.venv) root@sl2619:~/examples# ls /usr/share/synap/models/
image_classification object_detection placeholder

ls /usr/share/synap/models/object_detection/coco/
cpu/ info.json npu/

(.venv) root@sl2619:~/examples# ls /usr/share/synap/models/image_classification/
mobilenetv2

How to get access to the models that will run correctly? Can you please point out to the correct model/repo?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions