what's y_cls & y_cls[i,T]?

[MuseCP]

can someone show me the DataLoader? I don't know how to creat y_cls

[zhuruzhao]

@MuseCP have you worked it out? I have no idea how to create y_cls too

[MuseCP]

yes you need write a dataloeader yourself. return values are input，groundtruth，and a 1xn tensor contains class imformation。n is class num，1 is the label has the class and 0 is not | | 翟墨 邮箱：[email protected] | 签名由 网易邮箱大师 定制 On 04/01/2019 19:27, 1535315854 wrote: @MuseCP have you worked it out? I have no idea how to create y_cls too — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[zhuruzhao]

@MuseCP Thanks, I have written the dataloader yesterday. I also changed a few lines to make the code run. The problem I am facing now is how to evaluate the results? Do I have to write the evaluate code by myself too?

[MuseCP]

yes I write the code myself too | | 翟墨 邮箱：[email protected] | 签名由 网易邮箱大师 定制 On 04/02/2019 10:14, 1535315854 wrote: @MuseCP Thanks, I have written the dataloader yesterday. I also changed a few lines to make the code run. The problem I am facing now is how to evaluate the results? Do I have to write the evaluate code by myself too? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.

[zhuruzhao]

@MuseCP So what's your model performance? Could you please share your evaluate code? Thanks a lot!

[MuseCP]

i did not write the evaluate part，only output the final result | | 翟墨 邮箱：[email protected] | 签名由 网易邮箱大师 定制 On 04/02/2019 10:25, 1535315854 wrote: @MuseCP So what's your model performance? Could you please share your evaluate code? Thanks a lot! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or mute the thread.