NVIDIA-Merlin / NVIDIA-Merlin/Merlin
[QST] out of memory error while trying out examples in jupyter notebook
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 907
- Forks
- 129
- PR merge metrics
- No merged PRs in 30d
Description
❓ 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:
train_dataset = nvt.Dataset([os.path.join(INPUT_DATA_DIR, "train.parquet")])
valid_dataset = nvt.Dataset([os.path.join(INPUT_DATA_DIR, "valid.parquet")])
%%time
workflow.fit(train_dataset)
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with 02-ETL-with-NVTabular.ipynb and the workflow.fit(train_dataset) call in the reported Ubuntu, Docker, and GPU environment. Reproduce the CUDA out-of-memory error and determine the documented change or example adjustment needed for the notebook to run successfully.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- docker, jupyter-notebook, python
- Domain
- data-engineering, machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100