Closed Jingchensun closed 1 year ago
Hi @Jingchensun , I recommend checking this post: https://github.com/dusty-nv/jetson-inference/issues/724#issuecomment-694832441
Segmentation networks are heavy and the datasets typically large, so may need to train on PC / server with discrete GPU instead. If you use a small custom dataset like in that post I linked to, you may be able to do it on Jetson though.
Hi @Jingchensun , I recommend checking this post: #724 (comment)
Segmentation networks are heavy and the datasets typically large, so may need to train on PC / server with discrete GPU instead. If you use a small custom dataset like in that post I linked to, you may be able to do it on Jetson though.
Hi, thank you for your suggestion, I re-trained the segmentation model on my server successfully with your instruction.
Hi,Can you teach me how to do that,thanks! @Jingchensun
Please what are the steps to retrain the segnet model
Hi @Yhaw, please see this post - https://www.highvoltagecode.com/post/edge-ai-semantic-segmentation-on-nvidia-jetson
Thank you
On Sun, Mar 26, 2023 at 5:41 AM Dustin Franklin @.***> wrote:
Hi @Yhaw https://github.com/Yhaw, please see this post - https://www.highvoltagecode.com/post/edge-ai-semantic-segmentation-on-nvidia-jetson
— Reply to this email directly, view it on GitHub https://github.com/dusty-nv/jetson-inference/issues/730#issuecomment-1484001663, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMOOJ4PHDRUZ5G4H5HXJZ7TW57JJTANCNFSM4RS35TKQ . You are receiving this because you were mentioned.Message ID: @.***>
I saw you have released Classification/Recognition (ResNet-18) and Object Detection (SSD-Mobilenet), but how about re-training of Segmentation Networks, I want use the segnet to train my own dataset, how should I do?