load model issue

Question

load model issue

Closed this issue 5 years ago · 7 comments

Thank you for your excellent work.
Just like issue 5, I ran into the same problem and added "strict = False" to the solution you gave, but the problem still hasn't been solved and the same error still occurs. I tried to load res2net-50-26w-4s.
Look forward to your reply.
cuda：8.0
python：3.6
pytorch：0.4.0
torchvision：0.1.18

Answer 1 · 2020-02-24T16:59:22.000Z

Have you tried to run 'python res2net.py' directly? Or you can share with me an example code, so I can check where the problem is.

Answer 2 · 2020-02-24T17:02:02.000Z

#5 (comment)
maybe this solution can help you.

Answer 3 · 2020-02-24T17:05:32.000Z

#5 (comment)
maybe this solution can help you.

I try to solve this problem, but the same mistake will occur.

Answer 4 · 2020-02-24T17:06:35.000Z

XiaoHao-Chen commented 5 years ago

Answer 5 · 2020-02-24T17:12:47.000Z

It seems that torch0.4.1 is a little different with torch0.4.0.
You can upgrade from torch0.4.0 to torch0.4.1 or higher version to solve this problem.
ref:
https://discuss.pytorch.org/t/solved-unexpected-key-s-in-state-dict-batches-tracked/23884

Answer 6 · 2020-02-24T17:15:27.000Z

When I run 'python res2net.py' directly,the same issue will occur. I will try to upgrade the pytorch.
Thank you for your prompt reply.

Answer 7 · 2020-02-25T19:11:02.000Z

Updating the version of pytorch can solve the problem.
Thank you.