search

hitzhangyu / Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only

Self-supervised Image Enhancement Network: Training With Low Light Images Only
137 stars 18 forks source link

关于TF版本的问题 #9

Open hongbojia opened 3 years ago

hongbojia commented 3 years ago

您好，打扰您了，想问一下你这个环境是TF的那个版本，麻烦您注明下，谢谢

hitzhangyu commented 3 years ago

在我的电脑上，1.14.0和2.2.0 都可以运行

Gera001 commented 3 years ago

1.8的报道

hitzhangyu commented 3 years ago

啊，啥意思？

Gera001 commented 3 years ago

哈哈，就是我用tf1.8的也跑出来了，谢谢谢谢

------------------ 原始邮件 ------------------ 发件人: "hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only" @.>; 发送时间: 2021年4月12日(星期一) 上午10:59 @.>; @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9)

啊，啥意思？

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

Gera001 commented 3 years ago

老兄还知道用gan的方法怎么做吗，谢谢

---原始邮件--- 发件人: @.> 发送时间: 2021年4月12日(周一) 上午10:59 收件人: @.>; 抄送: @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9)

啊，啥意思？

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

hitzhangyu commented 3 years ago

我后边做的一个无监督+去噪版本的：https://github.com/hitzhangyu/V2-Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only Gan的方法可以看：EnlightenGAN 无监督的可以看：Zero-DCE

Gera001 commented 3 years ago

哈哈，我看过啦，优化了一下，但是那个enlightengan我好像跑不通

---原始邮件--- 发件人: @.> 发送时间: 2021年4月12日(周一) 中午11:11 收件人: @.>; 抄送: @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9)

我后边做的一个无监督+去噪版本的：https://github.com/hitzhangyu/V2-Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only Gan的方法可以看：EnlightenGAN 无监督的可以看：Zero-DCE

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

hitzhangyu commented 3 years ago

可以仔细看下他们的输入数据那部分的code，应该是你输入数据的文件夹和他们code里边写的不一致带来的

Gera001 commented 3 years ago

testA和testB这两个文件夹都需要图片吗

---原始邮件--- 发件人: @.> 发送时间: 2021年4月12日(周一) 中午11:16 收件人: @.>; 抄送: @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9)

可以仔细看下他们的输入数据那部分的code，应该是你输入数据的文件夹和他们code里边写的不一致带来的

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

hitzhangyu commented 3 years ago

这个忘了，我记得需要改的就这一部分，改完就可以跑了

Gera001 commented 3 years ago

学长是用GPU跑的吗

---原始邮件--- 发件人: @.> 发送时间: 2021年4月12日(周一) 中午11:26 收件人: @.>; 抄送: @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9)

这个忘了，我记得需要改的就这一部分，改完就可以跑了

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

hitzhangyu commented 3 years ago

对啊，cpu太慢了啊

Gera001 commented 3 years ago

好的，实验条件有限，没有GPU哈哈，麻烦学长了，谢谢学长解答

---原始邮件--- 发件人: @.> 发送时间: 2021年4月12日(周一) 中午11:37 收件人: @.>; 抄送: @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9)

对啊，cpu太慢了啊

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

Gera001 commented 3 years ago

想用enlightengan直接测试一下

---原始邮件--- 发件人: @.> 发送时间: 2021年4月12日(周一) 中午11:37 收件人: @.>; 抄送: @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9)

对啊，cpu太慢了啊

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

Zruto commented 3 years ago

想用enlightengan直接测试一下 … ---原始邮件--- 发件人: @.> 发送时间: 2021年4月12日(周一) 中午11:37 收件人: @.>; 抄送: @.**@.>; 主题: Re: [hitzhangyu/Self-supervised-Image-Enhancement-Network-Training-With-Low-Light-Images-Only] 关于TF版本的问题 (#9) 对啊，cpu太慢了啊 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

testA 放你想测试的图片，testB随便放一张图片就可以了

Mornnn commented 10 months ago

你好 用的tensorflow==2.2.0，但是出现下面问题： Traceback (most recent call last): File "main.py", line 142, in tf.app.run() AttributeError: module 'tensorflow' has no attribute 'app'

Mornnn commented 10 months ago

你好 用的tensorflow==2.2.0，但是出现下面问题： Traceback (most recent call last): File "main.py", line 142, in tf.app.run() AttributeError: module 'tensorflow' has no attribute 'app'

python 3.7

hitzhangyu commented 10 months ago

https://blog.csdn.net/qq_24163555/article/details/102954704

Mornnn commented 10 months ago

https://blog.csdn.net/qq_24163555/article/details/102954704

恩恩这个解决啦～