search

autonomousvision / LaRa

[ECCV 2024] Efficient Large-Baseline Radiance Fields, a feed-forward 2DGS model
https://apchenstu.github.io/LaRa/
MIT License
206 stars 6 forks source link

Memory requirements #2

Open yejr0229 opened 2 weeks ago

yejr0229 commented 2 weeks ago

Hi, I'd like to know how much memory is required for LaRa? An A100 or just RTX3090 with 24GB will be enough?

apchenstu commented 2 weeks ago

Great question. I used 4 A100-40G GPUs with a batch size of 3. I haven't tried training on a single GPU, but I believe the current implementation supports small batch sizes such as 1 or 2.

yosun commented 2 weeks ago

yejr - did you get it to work on RTX3090