Issue with saving the best model state

[yuanx749]

Issue description

Hi, thanks for this great tool!

I believe the line below should be self.best_model_dict = deepcopy(self.model.state_dict()). Otherwise the best model keeps getting updated, because model.state_dict() returns a reference to the dict.

PyPOTS/pypots/base.py

Line 782 in 187852a

self.best_model_dict = self.model.state_dict()

I encountered this issue when training a model with more epochs, I got different results even though the best model is at the same epoch.

Your contribution

As suggested in the description.

[github-actions]

Hi there 👋,

Thank you so much for your attention to PyPOTS! You can follow me on GitHub to receive the latest news of PyPOTS.
If you find PyPOTS helpful to your work, please star⭐️ this repository to get this issue prioritized.
Your star is your recognition, which can help more people notice PyPOTS and grow PyPOTS community.
It matters and is definitely a kind of contribution to the community.

Best,
Wenjie

[WenjieDu]

Hi @yuanx749, thanks for raising this issue. I agree with you. I'm going to label this issue as bug. Would you like to make a PR to fix this? This could be your first contribution to PyPOTS ;-)

[yuanx749]

Hi @WenjieDu , thanks for your quick response. I'll look into this next week, but since it's a small change, you can go ahead and make the fix if you'd like. :)

[WenjieDu]

No problem, Xiao. I've created PR #668 to fix it. Please have a review ;~)