v0.11.0

Dec 13, 2025

1. Support plateau the learning rate scheduler in qtcls/schedulers/__init__.py.
2. Add the qtcls/dataloaders subpackage and provide a mini implementation MiniDataLoader in mini_loader.py.
3. Provide more coding tips for working with config files (see qtcls/optimizers/__init__.py and qtcls/schedulers/__init__.py).
4. Add a condition to fix the division by zero issue when calling meter.global_avg.
5. Add csv_loader and csv_saver to qtcls/utils/io.py.
6. Update the demo criterion code in qtcls/criterions/_demo_.py and docs.
7. Optimize the hints logic when not using pre-training.
8. Minor adjustments.