trainer.test() with given checkpoint logs last epoch instead of checkpoint epoch · Lightning-AI/pytorch-lightning#20052

(1 comment) (1 reaction) (0 assignees)Python (26,687 stars) (3,233 forks)batch import

bughelp wantedrepro needed

説明

Bug description

Testing from a given checkpoint leads to logging the epoch number of the last checkpoint instead of the checkpoint specified:

trainer = Trainer(..., max_epochs=10)
lightning_module = MyLightningModule(...)
datamodule = MyDatamodule()

trainer.fit(lightning_module , datamodule=datamodule)

trainer.test(lightning_module , datamodule=datamodule, ckpt_path="last")     # <-- ok: logs correct epoch and step
ckpt_path="/.../checkpoints/epoch=2-step=396.ckpt"
trainer.test(lightning_module , datamodule=datamodule, ckpt_path=ckpt_path)  # <-- incorrect: logs last epoch and step

The second test logs epoch 10 instead of epoch 2. Similarly, the step number of the second test is incorrect.

What version are you seeing the problem on?

v2.2.1

コントリビューターガイド

技術スタック: python
領域: backendobservability
Issue 種別: bug
難度: 3
推定時間: 1-3 hours
活動状況: fresh
明確さ: clear
前提条件: Basic knowledge of PyTorch LightningUnderstanding of checkpoint loading
初心者向け度: 70
調査方針: The bug is in the trainer.test() method where logging uses the global epoch/step instead of the checkpoint's epoch/step. Likely the fix is in pytorch lightning/trainer/trainer.py or the logging callback (pytorch lightning/callbacks/model checkpoint.py). Check how ckpt path is used to restore state and ensure that the logger's current epoch and global step are set from the checkpoint. The issue has clear reproduction steps and no assignees, making it a good candidate for a first time contributor.