search

devendrachaplot / Object-Goal-Navigation

Pytorch code for NeurIPS-20 Paper "Object Goal Navigation using Goal-Oriented Semantic Exploration"
https://devendrachaplot.github.io/projects/semantic-exploration
MIT License
323 stars 59 forks source link

Why is this module "Semantic Mapping "not trained? #28

Open Sky-railway opened 7 months ago

Sky-railway commented 7 months ago

Thanks for the work. Doesn't this module need training? in the main.py file:

Semantic Mapping

sem_map_module = Semantic_Mapping(args).to(device)
sem_map_module.eval()
wuyukun-tong commented 7 months ago

Hi I want to known what gpu you use to run this code, I tried to reproduce this code on 3090 but report "CUDA out of memory". I hope your question i solved as well!

zyjzxy9 commented 7 months ago

This is the mean that your total memory is not enough for running the code. what is the total memory of your 3090 GPU? My 3090 for running this code is equipped with dual 3090 graphics cards, each having 24GB memory. you can try to reduce the number of threads to run this code successfully.

wuyukun-tong commented 6 months ago

thanks for your replay

---- Replied Message ---- | From | @.> | | Date | 04/06/2024 18:16 | | To | @.> | | Cc | @.>@.> | | Subject | Re: [devendrachaplot/Object-Goal-Navigation] Why is this module "Semantic Mapping "not trained? (Issue #28) |

This is the mean that your total memory is not enough for running the code. what is the total memory of your 3090 GPU? My 3090 for running this code is equipped with dual 3090 graphics cards, each having 24GB memory. you can try to reduce the number of threads to run this code successfully.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: @.***>