CUDA out of memory error

[Anji-Builds]

Thanks a lot for your interesting work!!

I am trying to run the code in ubuntu 22.04 using GeForce GTX TITAN X (12gb) and i am getting out of memory error. Could someone suggest how I can solve this?

Also if the batch size needs to be changed, which file should I find the variable in?

I got this error when for scene id 114 and no other changes.

Got this error after I changed the ray_batch_size from 8192 to 4096.

I am new to this, please need guidance!

[Mayakshamanjirinesht]

@Anji-Builds Hi, I am also facing the same issue, when i am done training, and evaluation starts. My GPU memory is 8GB, so how can we change the batch size? is changing the ray_batch_size same as changing the batch size?
Also how to do the inference just on camera images alone?

[Anji-Builds]

@Mayakshamanjirinesht No, I have not got a solution and even I am confused about the same things.

[Mayakshanesht]

@Anji-Builds can we connect by email at [email protected] or maybe on LinkedIn (https://www.linkedin.com/in/mayur-waghchoure-a5aba5ab?utm_source=share&utm_campaign=share_via&utm_content=profile&utm_medium=android_app)
To discuss this emernerf architecture, I am using it as a backbone. I trained it for the 1000 epochs but when I ran it in the eval model, the state dict of the saved checkpoint and new model which is supposed to adjust to the new scene had different sizes.