no file named params_no_priori.pkl

[gittigxuy]

PATH_MODEL_PARAMS = './params/params_no_priori.pkl' where could I find this file?

[James-Yip]

Sorry, I didn't upload the trained parameters since it's about 1GB and my result is not good enough as the paper.

[gittigxuy]

I just want to test this repo，could you please send this file to me？my email [email protected]，or you could share this file by baidu netdisk，thanks

[James-Yip]

@gittigxuy 链接: https://pan.baidu.com/s/1K0X287ingEEQtDn8zoCThQ 密码: 7mrh

[gittigxuy]

how about the pytorch version,does it support pytorch0.4?

[James-Yip]

Yes. This is a pytorch0.4 implementation.

[xuezu29]

Hi, could you please send 'params_no_priori.pkl' to me？my email is [email protected]，the baidu netdisk links have expired or share baidu netdisk links again, thanks a lot.

[xinyaojiejie]

Hi , could you share me the baidu disk link for the pretrained model? I click the link but it says that the sharing is outdated. My email is : [email protected]. Thanks a lot!!

[noprohuangqi]

can you share this file again? Including baidu disk.Thanks.

[James-Yip]

@noprohuangqi 链接: https://pan.baidu.com/s/15Br7Hhn_no-WBe1D5kXNSA 提取码: y8an

[noprohuangqi]

@noprohuangqi 链接: https://pan.baidu.com/s/15Br7Hhn_no-WBe1D5kXNSA 提取码: y8an

Thanks for your reply. Can I have a talk with you about this paper. My wechat number is hq2365528.

[minicraby]

@noprohuangqi 链接: https://pan.baidu.com/s/15Br7Hhn_no-WBe1D5kXNSA 提取码: y8an

Thanks for your share. Could you please help to figure it out that how is the file generated and what is it for? If I'd like to run the code with my own dataset instead of those in the paper,should I generate a new pkl file for it?

[James-Yip]

@noprohuangqi 链接: https://pan.baidu.com/s/15Br7Hhn_no-WBe1D5kXNSA 提取码: y8an

Thanks for your share. Could you please help to figure it out that how is the file generated and what is it for? If I'd like to run the code with my own dataset instead of those in the paper,should I generate a new pkl file for it?

There're trained parameters inside the pkl file generated by torch.save. To train the model on your own dataset, you need to modify the Dataset part in 'main.py', and run python train.py in command line. After training, you could replace the constant PATH_MODEL_PARAMS in main.py with your own pkl file path for further training or testing. Hope this helps!