Jumponthemoon

Results 16 comments of Jumponthemoon

Btw the mobilenet setting as get_mobile_net(5,{'hm':18,'reg':2,'wh':2}),the head_conv set as default 64. I'm not sure if the head_conv number is too large or should i set the head_conv as 24

Hi, Thanks for your reply! I set the MODEL.CLOUDS.WEIGHTS to be the warm-up phase trained model, but here is the issue i encountered: ![image](https://github.com/yasserben/CLOUDS/assets/39290403/7333c069-6b8c-4136-976b-2a3d377f72e1) I don't know what's wrong. I'm...

Hi, The saved checkpoint name is 'sem_seg_evaluation.pth' after warm-up training. So i set it to be: ![image](https://github.com/yasserben/CLOUDS/assets/39290403/13531f76-a840-4ea9-9647-ff6e2a6c9de0)

> yeah I think I'll choose to move onto 3dgs. still really appreciate your kind and detailed response! Hi, i'm wondering if you can successfully run on single RTX 4090?

> I tested with my 4090, it is okay Thank you so much!

Me the same. I can train with batch size 1 and image size 256x256 by using RTX A6000 50GB memory. It consumes 26GB memory. When i try to increase resolution...