Lightning-AI/pytorch-lightning

trainer.test() with given checkpoint logs last epoch instead of checkpoint epoch

Open

#20,052 建立於 2024年7月5日

在 GitHub 查看
 (1 留言) (1 反應) (0 負責人)Python (26,687 star) (3,233 fork)batch import
bughelp wantedrepro needed

描述

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

貢獻者指南

技術棧
python
領域
backendobservability
議題類型
bug
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
3
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
1-3 hours
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
fresh
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
clear
前置要求
Basic knowledge of PyTorch LightningUnderstanding of checkpoint loading
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
70
研究方向
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.
trainer.test() with given checkpoint logs last epoch instead of checkpoint epoch · Lightning-AI/pytorch-lightning#20052 | Good First Issue