about the results of training

[feidasheng]

I would like to ask if your code is about the generated result image. I trained only the .mat file according to your main file, and there is no specific visualization picture? such as your github image

[Hans1984]

There is a simple visualization function in the test script which is implemented by matplotlib.

[feidasheng]

sorry，i  do not find your test script, i only find demo ,main and tool python code, can you tell me where the script test , thank you very much !

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "Hans1984"<notifications@github.com&gt;; 发送时间: 2020年1月17日(星期五) 下午2:48 收件人: "Yang7879/3D-RecGAN-extended"<3D-RecGAN-extended@noreply.github.com&gt;; 抄送: "╭学絵パ紾悕"<1170776918@qq.com&gt;; "Author"<author@noreply.github.com&gt;; 主题: Re: [Yang7879/3D-RecGAN-extended] about the results of training (#13) There is a simple visualization function in the test script which is implemented by matplotlib. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[Hans1984]

Here, https://github.com/Yang7879/3D-RecGAN-extended/blob/master/demo_3D-RecGAN%2B%2B.py
and you can use the visualize() function.

[feidasheng]

hello，this demo .py i run meet this problem "cannot feed value of shape (1,64,64,64,1) for tensor 'Placeholder:0',which has shape '(4,64,64,64.1)'" ,i find other people also meet this problem because i pip different tensorflow :1.13.1,you publish the version of tensorflow is too old , cannot match our cuda toolkit, you think how we can slove this problem? &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;thank you very much! &nbsp;

…

------------------&nbsp;原始邮件&nbsp;------------------ 发件人: "Hans1984"<notifications@github.com&gt;; 发送时间: 2020年1月17日(星期五) 下午3:19 收件人: "Yang7879/3D-RecGAN-extended"<3D-RecGAN-extended@noreply.github.com&gt;; 抄送: "╭学絵パ紾悕"<1170776918@qq.com&gt;; "Author"<author@noreply.github.com&gt;; 主题: Re: [Yang7879/3D-RecGAN-extended] about the results of training (#13) Here, https://github.com/Yang7879/3D-RecGAN-extended/blob/master/demo_3D-RecGAN%2B%2B.py and you can use the visualize() function. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.

[Hans1984]

maybe you can degrade the version of tensoflow

[jinzita142857]

@feidasheng Hello, I also encountered such a problem, how did you solve it effectively? Lowering the version?