Pretrain, finetune ANY AI model of ANY size on 1 or 10,000+ GPUs with zero code changes.
Fix LightningDataModule.load_from_checkpoint to restore subclass and hyperparameters (#21478)
* add args and kwargs to Data module constructor * test for load_from_checkpoint for data module * remove additional params passed to superclasses * merge test into *injection_subclass_mode test * add chlog --------- Co-authored-by: bhimrazy <bhimrajyadav977@gmail.com>
A
Aditya Khedekar committed
f2da71fb46b6ecba70a00f46cbb3ffca9ad8601d
Parent: 2060299
Committed by GitHub <noreply@github.com>
on 1/12/2026, 10:00:29 AM