-
Notifications
You must be signed in to change notification settings - Fork 111
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[QST] out of memory error while trying out examples in jupyter notebook #993
Comments
@Bartoliinii hello. Can you clarify the followings:
Besides, the docker image version you mentioned above is like one year old. Is it possible for you to use our latest stable docker image? Can you please test |
Hi @rnyak thanks for responding.
It was this notebook: https://nvidia-merlin.github.io/NVTabular/v0.7.1/examples/getting-started-movielens/02-ETL-with-NVTabular.html
I was running cells, with no custom inputs on my part
I was running it as a docker image
I know, but the last time I tried to use a newer docker image with tensorflow I had to update drivers for nvidia graphics card, this update wrecked my cuda installation.
|
you can directly read data from path to NVT.. like that:
does that answer your question? btw, do you have 8 GB machine? |
Yes, thank you.
Yes, I do. |
@Bartoliinii I noticed that you are following examples in the NVT website.. Please use these notebooks instead: https://github.com/NVIDIA-Merlin/Merlin/tree/main/examples/getting-started-movielens thanks. |
Thank you for the tip. I'm following the instructions in this docker container for TensorFlow:
In the Cell 13:
Cell 14:
Output: Is there a way to fit workflow to a dataset without filling GPU memmory? |
so you can change all
Note that
the moment you do
Please be sure you are not using anything like |
Unfortunately, the error still remains. Here is the full output:
|
❓ Questions & Help
Details
Hi, I have experienced CUDA out of memory error while trying out examples in jupyter notebook files. Is there any neat way of solving this error?
os: Ubuntu 20.04
gpu: NVIDIA GeForce RTX 3060 Ti
cuda: 11.8
cudnn: 8
docker version: 24.0.1
docker image: nvcr.io/nvidia/merlin/merlin-pytorch-training:22.03
ipynb file: 02-ETL-with-NVTabular.ipynb
The code that throws an error
cudaErrorMemoryAllocation out of memory
:The text was updated successfully, but these errors were encountered: