Project-MONAI / Project-MONAI/MONAI
Auto3d Caches data depending on number of gpus.
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 8.7k
- Forks
- 1.6k
- Avg merge
- 5d 1h
- Merged PRs (30d)
- 20
Description
Describe the bug
Cache rate is dependent on the size of the data and the system memory. Current code
changes it depending on the number of gpus used.
Please change that logic and make an option for user to explicit overwrite that parameter or it should be a function os system memory size and not number of gpus
Related issue Please place back the caching visualization to indicate data being cached. For large data it seems like the systems hanged up
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 auto3dseg/algorithm_templates/dints/scripts/train.py, especially lines 133-154, and trace how cache sizing is calculated from GPU count, data size, and system memory. Done means cache sizing is not tied to GPU count, users can explicitly override it or it uses system memory, and the caching visualization is restored to show progress for large datasets.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 38/100