leftthomas / ESPCN

A PyTorch implementation of ESPCN based on CVPR 2016 paper "Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Could re-release training data or pre-train model again?

intere2960 opened this issue · comments

commented

As the title, all baidu yun url are missing, and I try to use pre-train model but have "UnpicklingError: invalid load key, '\xef'" problem when loading model. Could you release those data again? Or could you send those data to my email intere2960@yahoo.com.tw?

p.s. My platform is windows and use python 3.6.2 & pytorch 0.4.0. (windows don't have pytorch 0.3.1)

this code should be run on Ubuntu, windows load path will encounter error

commented

Unfortunately, I try another Ubuntu platform with python 3.6.8 & pytorch 0.3.1 still get "UnpicklingError: invalid load key, '\xef'". Could you help me to resolve my problem? I check all close issue on this respository but those solution can't work.

1
2

@intere2960 does these data links are missing? Train, Test.
And because I haven't save the pre-train model, so if it has been damaged, you should train from scratch by yourself, sorry for that.

commented

It seems those link can't access it. (PC & Mobile) Or I have to use VPN to access China internet domain?

PC:
image

Mobile:
Screenshot_20190406-165844

@intere2960 look at this repo. Are these links unavailable? The data is same with that repo, you could download from that repo.

commented

I get the same result as ESPCN. Can't access it.

@intere2960 I have updated these links

commented

I still can't access it, but I try VPN use japan, hong kong, US IP can access link. I don't know whether my IP address can't access baidu. (I'm in Taiwan)

@intere2960 I didn't know why, maybe Baidu restricts this service in some areas. Now these links in my repo are all right (I have tested), then if you can't access still, maybe you should change your IP address, or contact Baidu to ask help.