1回答
- 新しい順
- 投票が多い順
- コメントが多い順
0
Hi. This blog covers deploying yolov8 models to Greengrass, so it should point you in the right direction: https://aws.amazon.com/blogs/iot/deploying-and-benchmarking-yolov8-on-gpu-based-edge-devices-using-aws-iot-greengrass/
Associated repo: https://github.com/aws-samples/deploy-yolov8-on-edge-using-aws-iot-greengrass
Many thanks Greg, it is certainly very useful. The issue here is that in this tutorial you need to upload the model on the jetson but what I need is the ability to transfer my own model from aws to all edge machines in my fleet. Basically, is there any way to transfer my model save at S3 to every machine in my fleet? Thanks again
Once you have created GG component as explained in the blog, then you associate a deployment https://docs.aws.amazon.com/greengrass/v2/developerguide/create-deployments.html to deploy the components on the target greengrass group to make inference on all your devices https://docs.aws.amazon.com/greengrass/v2/developerguide/perform-machine-learning-inference.html. Does it make sense?