Pytorch version issue & Pretrained weight

[Sanghyun Woo]

Hello,

First of all, thanks for sharing your code.

I have several issues regarding to pytorch version and pretrained weight sharing request.

Due to dependency problem of another code, I am using now pytorch 0.4.0 version not 0.4.1 which is noted as a required version for this repo.

However, I found that by replacing F.interpolate function with F.upsample function makes everything fine. Is it okay to use this repo as this way?

Moreover, could you please share your pretrained weight ?

Looking forward to your reply.

Thanks.

[Naoto Inoue]

Using F.interpolate will be OK if the actual operation is doing nearest-neigbour sampling, be careful.
Here is the pre-trained weight, but I cannot guarantee the quality. If you find any bugs, please report them!

[sunjunlishi]

it could be not reachead？

[Naoto Inoue]

It's too large to be shown in the browser, but you can download it, don't you?

[sunjunlishi]

i have downloaded.My project is in need of watermarking and recognition

[peachis]

It seems like the link of the pre-train model you shared is unreachable. Would you please renew it.

Thanks a lot!

[Naoto Inoue]

I have updated here

[peachis]

Got it!
I have just tested it on Places2, and it works well.

Thank you again!