search

mbzuai-oryx / Video-ChatGPT

[ACL 2024 🔥] Video-ChatGPT is a video conversation model capable of generating meaningful conversation about videos. It combines the capabilities of LLMs with a pretrained visual encoder adapted for spatiotemporal video representation. We also introduce a rigorous 'Quantitative Evaluation Benchmarking' for video-based conversational models.
https://mbzuai-oryx.github.io/Video-ChatGPT
Creative Commons Attribution 4.0 International
1.23k stars 108 forks source link

Training GPU Device minimum Require Specs. #32

Closed BigJoon closed 1 year ago

BigJoon commented 1 year ago

Thanks for your wonderful work.

I saw you used 8 A100 40GB GPUs.

Is it possible to learn with 4 x RTX A6000 GPUs?

mmaaz60 commented 1 year ago

Hi @BigJoon,

Thank you for your interest in our work.

Training on 4 x RTX A6000 GPUs is possible. Please set the --gradient_accumulation_steps 2 to match the overall batch size used in our experiments.

Good Luck!

BigJoon commented 1 year ago

Thanks for your reply, @mmaaz60

I think this work is very interesting. I'll try to make a contribution someday.

Xinxinatg commented 1 year ago

Hi @BigJoon,

Thank you for your interest in our work.

Training on 4 x RTX A6000 GPUs is possible. Please set the --gradient_accumulation_steps 2 to match the overall batch size used in our experiments.

Good Luck!

would 2 A30 be able to support the training?

hkuit commented 1 year ago

Thanks for your wonderful work.

Is it possible to learn with 1 A100 80 GPUs? Thanks