Open dataset Updated 2026-06-17
What is debug?
The debug dataset is a small collection of records sized under 1K entries, created by rtrm for the ai-ml domain.
It supports basic testing scenarios where limited data volume is required.
What you can build with debug
Pipeline testing
Run data loading and preprocessing scripts against a tiny dataset to verify code works before scaling to larger data.
Model debugging
Feed the records into training loops to isolate bugs in model architecture or loss functions without long wait times.
CI/CD validation
Include the dataset in automated tests to confirm ML workflows execute correctly on every code commit.
Load debug
Python
from datasets import load_dataset
ds = load_dataset("rtrm/debug")- 1pip install datasets
- 2from datasets import load_dataset
- 3dataset = load_dataset('rtrm/debug')
- 4print(dataset) to inspect splits and features
- 5Access records via dataset['train'][0]
debug: pros & cons
Pros
- +Very small size enables instant loading and iteration
- +Directly available through Hugging Face datasets library
- +Simple structure with no extra dependencies
- +Useful for quick sanity checks during development
Cons
- –Fewer than 1K records limits statistical experiments
- –No tasks or labels documented
- –Appears to be test data only (source note: test3)
Did you find this helpful?
Frequently asked questions
A small test collection of fewer than 1K records created by rtrm for the ai-ml category.
User reviews
Verified reviews from the community shape this listing's rating.
Loading reviews…
Similar datasets
Other ai & machine learning options worth comparing.