Pretrained model?

[yuskey]

Do you have any pretrained model weights? I currently can't train something like this so was curious if you had anything pretrained available.

[taki0112]

I'm talking to the company about whether it's okay to release a pre-trained model.
Please wait a little. Sorry.

[chaowentao]

I'm talking to the company about whether it's okay to release a pre-trained model.
Please wait a little. Sorry.

I train the model in my own dataset. The result looks not very well. Hopefully you share your pre-trained model @taki0112

[Ledarium]

We want to make anime, please

[tafseerahmed]

If you open a patreon or something, we can subscribe for your pre-trained model 🗡

[cpury]

Alternatively, it would be amazing if you could share the selfie2anime dataset.

[vulcanfk]

Alternatively, it would be amazing if you could share the selfie2anime dataset.

See issue #6

[kioyong]

Can't wait to want a pre-trained model~ please~

[hensat]

Let's hope the company will allow you to place the model

[hafniz]

It would be really helpful if you could release some existing model for our reference. Please~

[eddybogosian]

Guys just chill for a moment. Taki already said that they're talking to the company about it. It's been 2 days. Calm down and wait. They know that we want this, flooding won't help.

In the meantime, why don't you try something yourself? You can use Microsofts Azure or Amazon AWS to train this type of network. Maybe you can even come up with something better! Who knows right?

[tafseerahmed]

The thing we need to understand is that no one likes begging and pleading. These people have worked hard on something, and it's completely up to them if they choose to release their models or datasets. I appreciate the fact that they open-sourced their code. Personally, I wouldn't mind even paying for their models and dataset. In the meantime let's stop flooding this thread and wait for @taki0112 's response.

[mickae1]

If you don't want to share, and I can understand. You should create a website that offer the possibility to convert photo into anime. The website will become very popular. And you can get some money with the publicity

[thewaifuai]

I have published a pre-trained model for cat2dog on kaggle. Please let me know if you have any issues with it. I saved the results in this pdf so you can see what it looks like:
results.pdf I used the cat2dog dataset from DRIT.

It takes 4+ days to train cropped face dataset and 16+ days to train cropped body dataset on Nvidia GPUs (estimates). Since it takes many days to train the dataset once, and it takes many iterations of training it will take some time but eventually many people will publish and share their pre-trained models in the weeks to come. Datasets can be found at DRIT. For selfie2anime you can use datasets selfie and anime face dataset. Other potential anime face dataset sources: thiswaifudoesnotexist,
animeGAN for generating anime images and a one click download anime face dataset.
UGATIT is quite general, you really just need a folder of anime faces and a folder of human faces and it figures the rest by itself.

[amberjxd]

@thewaifuai Hello! Would you like to publish the pre-trained model of selfie2anime in future? Thanks.

[thewaifuai]

@thewaifuai Hello! Would you like to publish the pre-trained model of selfie2anime in future? Thanks.

Yes

[cpury]

FYI, I'm using a quickly-assembled, crappy dataset and a relatively slow cloud GPU machine. Also, I reduced the resolution to 100x100 pixels (256 just takes too long for me). The results look like this after one day of training:

Not too bad, but still a lot of room for improvement :)

What I can recommend if you'd like to create a better one:

• Make sure the two datasets have similar poses / distances to the face. You can tell in mine that the anime data is much more close-up to the face and so the model learned that part of the transformation is "zooming in".
• Make sure the anime dataset is diverse. Right now, in my model, everything from black men to old women gets transformed into 12-yo-looking girls with giant eyes, white skin, and bangs. I'd really rather it learns something more diverse...
• Get a serious cloud machine and expect to spend some time. The batch size of 1 is killing me 😅

[LiuShaohan]

FYI, I'm using a quickly-assembled, crappy dataset and a relatively slow cloud GPU machine. Also, I reduced the resolution to 100x100 pixels (256 just takes too long for me). The results look like this after one day of training:

Not too bad, but still a lot of room for improvement :)

What I can recommend if you'd like to create a better one:

• Make sure the two datasets have similar poses / distances to the face. You can tell in mine that the anime data is much more close-up to the face and so the model learned that part of the transformation is "zooming in".
• Make sure the anime dataset is diverse. Right now, in my model, everything from black men to old women gets transformed into 12-yo-looking girls with giant eyes, white skin, and bangs. I'd really rather it learns something more diverse...
• Get a serious cloud machine and expect to spend some time. The batch size of 1 is killing me 😅

FYI, I'm using a quickly-assembled, crappy dataset and a relatively slow cloud GPU machine. Also, I reduced the resolution to 100x100 pixels (256 just takes too long for me). The results look like this after one day of training:

Not too bad, but still a lot of room for improvement :)

What I can recommend if you'd like to create a better one:

• Make sure the two datasets have similar poses / distances to the face. You can tell in mine that the anime data is much more close-up to the face and so the model learned that part of the transformation is "zooming in".
• Make sure the anime dataset is diverse. Right now, in my model, everything from black men to old women gets transformed into 12-yo-looking girls with giant eyes, white skin, and bangs. I'd really rather it learns something more diverse...
• Get a serious cloud machine and expect to spend some time. The batch size of 1 is killing me 😅

Can you share your training dataset?

Or Pretrained model？

Thanks Very Much!~

This is my email：liushaohan001@gmail.com

[td0m]

I have published a pre-trained model for cat2dog on kaggle. Please let me know if you have any issues with it. I saved the results in this pdf so you can see what it looks like:
results.pdf I used the cat2dog dataset from DRIT.

@thewaifuai I'm not sure why but your cat2dog kaggle link doesn't work?

[thewaifuai]

I have published a pre-trained model for cat2dog on kaggle. Please let me know if you have any issues with it. I saved the results in this pdf so you can see what it looks like:
results.pdf I used the cat2dog dataset from DRIT.

@thewaifuai I'm not sure why but your cat2dog kaggle link doesn't work?

Oops kaggle datasets are private by default, I had to manually make it public. It is now public and should work.

[tafseerahmed]

I have published a pre-trained model for cat2dog on kaggle. Please let me know if you have any issues with it. I saved the results in this pdf so you can see what it looks like:
results.pdf I used the cat2dog dataset from DRIT.

I am actively working on writing a TPU version of UGATIT. If anyone is interested please respond to my UGATIT TPU issue. I am interested with working with others to make the TPU version.

It takes 4+ days to train cropped face dataset and 16+ days to train cropped body dataset on Nvidia GPUs (estimates). Since it takes many days to train the dataset once, and it takes many iterations of training it will take some time but eventually many people will publish and share their pre-trained models in the weeks to come. Datasets can be found at DRIT. For selfie2anime you can use datasets selfie and anime face dataset. Other potential anime face dataset sources: thiswaifudoesnotexist,
animeGAN for generating anime images and a one click download anime face dataset.
UGATIT is quite general, you really just need a folder of anime faces and a folder of human faces and it figures the rest by itself.

Should the images in trainA and trainB be of same sizes? the selfies are 306x306 but my anime faces were 512x512 mixed pngs and jpgs. I did run into some errors.

[tafseerahmed]

This is on a 4x P100 with 11 GB VRAM on each trainA is selfie dataset and trainB is http://www.seeprettyface.com/mydataset_page2.html + 1k dump of male anime from gwern's TWDNEv2 website.

I guess, if I reduce the batch size? then I can quickly train and release the pre-trained models.

[cpury]

@tafseerahmed the size and format of the images shouldn't matter. They get resized anyway AFAIK.

The error you're getting is OOM - out of memory. I believe you don't have enough available RAM (as opposed to GPU memory) to create the model. Is that possible?

[thewaifuai]

@tafseerahmed use the --light True option, if that does not work run pkill python3 and then try again with the --light True option. This runs the light version of UGATIT.

[tafseerahmed]

@tafseerahmed the size and format of the images shouldn't matter. They get resized anyway AFAIK.

The error you're getting is OOM - out of memory. I believe you don't have enough available RAM (as opposed to GPU memory) to create the model. Is that possible?

Someone is using 2 GPU's right now but I still have over 256GB of RAM available.

[tafseerahmed]

@tafseerahmed use the --light True option, if that does not work run pkill python3 and then try again with the --light True option. This runs the light version of UGATIT.

wouldn't that reduce the quality of final results?

[thewaifuai]

@tafseerahmed use the --light True option, if that does not work run pkill python3 and then try again with the --light True option. This runs the light version of UGATIT.

wouldn't that reduce the quality of final results?

Yes

[tafseerahmed]

@tafseerahmed use the --light True option, if that does not work run pkill python3 and then try again with the --light True option. This runs the light version of UGATIT.

wouldn't that reduce the quality of final results?

Yes

lol thanks its training now

but did you train yours on the heavy model instead of light? I imagine the full model requires more than 16GB VRAM

[cpury]

The light version significantly reduces the capacity of the model. I haven't trained for long but I don't think it's worth trying.

With that hardware, you really should not have any memory issues. Maybe the dataset is too big and already takes up most the memory? I don't know but I think you should investigate / experiment more.

[tafseerahmed]

The light version significantly reduces the capacity of the model. I haven't trained for long but I don't think it's worth trying.

With that hardware, you really should not have any memory issues. Maybe the dataset is too big and already takes up most the memory? I don't know but I think you should investigate / experiment more.

the batch size was set to 1 by default (that's ineffective when you have a GPU), so I can't imagine that the hardware was an issue. I will debug more and let you guys know, in the meantime, I am training on the light model.

[Gaggsta]

Tell me please. If I start training, will the model improve or will another model be created?

[onefish51]

1. We released 50 epoch and 100 epoch checkpoints so that people could test more widely.
2. Also, We published the selfie2anime datasets we used in the paper.
3. And, we fixed code in smoothing
4. In the test image, I recommend that your face be in the center.

===================
Sorry, you can't view or download this file at present.
Too many users have recently viewed or downloaded this file. Please try to access this file later. If you try to access a file that is particularly large or shared by many people, it may take up to 24 hours to view or download the file. If you are still unable to access the file after 24 hours, please contact your domain administrator.
抱歉，您目前无法查看或下载此文件。

最近查看或下载此文件的用户过多。请稍后再尝试访问此文件。如果您尝试访问的文件特别大或由很多人一起共享，则可能需要长达 24 小时才能查看或下载该文件。如果 24 小时后您仍然无法访问文件，请与您的域管理员联系。

50 epoch and 100 epoch checkpoints ，Who can share BaiduYun after downloading ? Thank you！
请问有谁能下载后分享到百度云盘吗？谢谢！
QQ2737499951共同研究

存到google drive后右键保存副本进行下载

谢谢，但是这个文件估计禁止复制副本了， 我复制副本出错，您能下载一个看看吗？谢谢您qq多少，加个学习一下，谢谢

看下这个
主要超4g了，度盘上传麻烦，要分卷。

I download the "selfie2anime checkpoint (100 epoch)" from google drive, upload it to baidu drive.

链接:https://pan.baidu.com/s/1e4jqW2ZcaE5P_OyhIusd6w 密码:5io9

[t04glovern]

We've created a web site that can be used to try the 100 epoch model out.

https://selfie2anime.com/

[leemengtw]

We've created a web site that can be used to try the 100 epoch model out.

@t04glovern Thanks for making this more reachable to the general public!

But because you're using the pre-trained model directly open source by the authors, maybe you guys would want to give more credits to the authors including @taki0112 and add some direct link to their GitHub repo and paper on the website. (there is no link to the original fantastic work by the authors right now).

[t04glovern]

We've created a web site that can be used to try the 100 epoch model out.

@t04glovern Thanks for making this more reachable to the general public!

But because you're using the pre-trained model directly open source by the authors, maybe you guys would want to give more credits to the authors including @taki0112 and add some direct link to their GitHub repo and paper on the website. (there is no link to the original fantastic work by the authors right now).

Hey @leemengtaiwan, sorry; we had the links in a non-cached version of cloudfront. If you refresh now they should be listed instead of our repo. Definitely don't want to undermine any of their amazing work!

[Losses]

@t04glovern Thanks! I finally got a chance to try this model!

Ohh... It's cyberpunk... 😂

[creke]

I also make a self2anime website using the official pretrain model.
Simple but optimized for mobile devices.
https://waifu.lofiu.com/

[sdy0803]

I also make a self2anime website using the official pretrain model.
Simple but optimized for mobile devices.
https://waifu.lofiu.com/

exciting!

[cpury]

I'm having trouble extracting it, too 🤔 If someone who managed to do so could re-archive the file as 7z or tar.gz or similar and re-upload it, that would be amazing!

[eqiihuu]

I also make a self2anime website using the official pretrain model.
Simple but optimized for mobile devices.
https://waifu.lofiu.com/

Hi Creke, it seems like your website works pretty good. Are you using the 100-epoch official pretrain model?

[leemengtw]

@eqiihuu I think one of the key reasons why @creke 's website work so well is because there is a face detection & crop step before applying the model. Not sure whether it is 100 or 50 epoch pre-trained model though :)

[eqiihuu]

@eqiihuu I think one of the key reasons why @creke 's website work so well is because there is a face detection & crop step before applying the model. Not sure whether it is 100 or 50 epoch pre-trained model though :)

Yeah, totally agree with you. That crop helps a lot.

[QQ2737499951]

UGATIT 预训练模型和数据集百度云下载，为方便谷歌硬盘无法使用的人，不懂的可加Q群264191384 注明UGATIT 共同研究 ，谢谢！

官方已经开放的数据集和模型在google硬盘，比较大，可能无法下载，我分卷压缩了一下，分享到百度云，方便大家下载。
Pretrained model： selfie2anime checkpoint (100 epoch)
Dataset：selfie2anime dataset

目前win10下的测试训练自己的数据问题，都可以 ，UGATIT 预训练模型和数据集百度云下载，为方便谷歌硬盘无法使用的人，不懂的可加Q群264191384 注明UGATIT 共同研究 ，谢谢！

链接：https://pan.baidu.com/s/1dP1mXuU-rA9dPvFe8YS8jQ
提取码：k6rc

> 已添加checkpoint文件

下载后合并01.02.03为一个文件。
目录结构如下：
B:\Tensorflow\UGATIT\checkpoint\UGATIT_selfie2anime_lsgan_4resblock_6dis_1_1_10_10_1000_sn_smoothing

=================

已添加checkpoint文件 直接下载解压即可

[Aneureka]

Hi there, I have built a wechat miniapp based on UGATIT and its 100-epoch pretrained model. Hope you will love it, and I will also add more features later.

Here's the screenshot 🤣

And the miniapp code:

The miniapp and back-end codes are open-source:

• https://github.com/Aneureka/animate-me
• https://github.com/Aneureka/UGATIT-s2a

It typically spends several seconds on inferring.

Thank you all!

[leemengtw]

@Aneureka I tried your app, well done!

[EndingCredits]

I trained my own model using the celebA dataset and a large dataset I found somewhere on here. Trained for 100 epochs.

It's more diverse than the existing model, but it can often give poor results. Also, unfortunately it seems to have picked up jpg artifacts.

It also goes the other way round reasonably well (but you have to match the art style properly).

Download here:
https://drive.google.com/open?id=1hrqPyy-skDarGqRektBy-G4sYYFjnMwm

Feel free to use in any web-apps or anything.

[cpury]

@EndingCredits that's amazing!! Thanks so much for sharing. A pity that it picked up the JPG-artifacts...

[abelghazinyan]

This is on a 4x P100 with 11 GB VRAM on each trainA is selfie dataset and trainB is http://www.seeprettyface.com/mydataset_page2.html + 1k dump of male anime from gwern's TWDNEv2 website.

I guess, if I reduce the batch size? then I can quickly train and release the pre-trained models.

How do you use multiple GPUs, because that feature is not implemented in the repo. I need it for training on my own dataset, and I am getting error from memory. I have 2x half k80 gpus 2x12gb Vram, but the model is using only one half.

[qweasdzxc110]

I trained my own model using the celebA dataset and a large dataset I found somewhere on here. Trained for 100 epochs.

It's more diverse than the existing model, but it can often give poor results. Also, unfortunately it seems to have picked up jpg artifacts.

It also goes the other way round reasonably well (but you have to match the art style properly).

Download here:
https://drive.google.com/open?id=1hrqPyy-skDarGqRektBy-G4sYYFjnMwm

Feel free to use in any web-apps or anything.

Amazing， can you tell that which anime datasets are picked???

[YangNaruto]

Hi there, I have built a wechat miniapp based on UGATIT and its 100-epoch pretrained model. Hope you will love it, and I will also add more features later.

Here's the screenshot 🤣

And the miniapp code:

The miniapp and back-end codes are open-source:

• https://github.com/Aneureka/animate-me
• https://github.com/Aneureka/UGATIT-s2a

It typically spends several seconds on inferring.

Thank you all!

thanks

[xyxxmb]

I trained my own model using the celebA dataset and a large dataset I found somewhere on here. Trained for 100 epochs.

It's more diverse than the existing model, but it can often give poor results. Also, unfortunately it seems to have picked up jpg artifacts.

It also goes the other way round reasonably well (but you have to match the art style properly).

Download here:
https://drive.google.com/open?id=1hrqPyy-skDarGqRektBy-G4sYYFjnMwm

Feel free to use in any web-apps or anything.

Looks Like better than the author. Could you tell me how get the anime (cartoon) dataset? Thank you.

[sankexin]

最近不知为啥不能科学上网，kaggle用电子邮件注册又验证不过...
哪位朋友下载了放在百度网盘上共享一下？
不敢发英文，怕给咱们国家的大局域网丢人...

同样不敢发英文，因为我英文太差 （QAQ）
我总结一下我看到的 数据集 和 预训练模型 :
ps：十分感谢他们的分享
来自heart4lor的 selfie2anime数据集 大小: 约110 MB
通过谷歌下载2*3500张训练集
通过百度下载密码：1exs
据他本人所说此数据集还有改进空间
来自t04glovern的 selfie2anime预训练模型
百度链接来自知乎 密码：50lt
他本人的回答，在他的分支中有resize.py工具来缩小图像大小
kaggle链接
这边提供他的数据集下载地址
人：crcv.ucf.edu/data/Selfie
动漫：gwern.net/Danbooru2018
来自thewaifuai的 cat2dog预训练模型
百度链接 密码：aw35
kaggle链接
或者你可以看看他的回答
cat2dog 数据集百度链接 密码：ryvj
他会把猫变成狗，反之亦然
来自知乎的 selfie2anime数据集
年轻女性-1000张-512px：百度链接 密码：udlm
二次元-1000张-512px：百度链接 密码：d1yg
以下为问答中看到的链接，我不确定他们是否对你有帮助：
http://www.seeprettyface.com/mydataset_page2.html
里面有 人 和 动漫 的高质量数据集，并且是通过 百度下载
这是一条 可以参考的问答 ，他提供了获取数据集的参考
似乎就这么多了
顺便一提，如果你的电脑没有足够的显存，我找到的方法是：
light = True，
调低 iteration 和 epoch？
缩小图像大小？
或者使用预训练模型测试
或者，可以等待更完善的模型？
希望对你有帮助

我训练了一个模型，然后想将其转换为.pb文件形式。不知道您了解这个输入输出节点是什么吗？

you can find in tensorboard

[EndingCredits]

I don't remeber exactly where I found the dataset, but it appears to be from Getchu.

If it's necessary, I can reupload, but I don't really want to do that unless it's a last resort.

[dragen1860]

Does anyone succeed to compress the model size and keep comparable performance. the whole size is up to 1GB, which really prevent UGATIT to be deployed widely!

[ghost]

@dragen1860 also interested in this were you able to find a solution, maybe https://github.com/mit-han-lab/gan-compression

[gdwei]

I trained my own model using the celebA dataset and a large dataset I found somewhere on here. Trained for 100 epochs.

It's more diverse than the existing model, but it can often give poor results. Also, unfortunately it seems to have picked up jpg artifacts.

It also goes the other way round reasonably well (but you have to match the art style properly).

Download here:
https://drive.google.com/open?id=1hrqPyy-skDarGqRektBy-G4sYYFjnMwm

Feel free to use in any web-apps or anything.

Well done! At my training, the discrimintator loss did not goes down, do you get any idea about it ?

[gdwei]

Some hand-picked results after two days of training on a low-quality dataset: https://twitter.com/cpury123/status/1159844171047301121

Well done! At my training, the discrimintator loss did not goes down, do you get any idea about it?

[calvin886]

I have published a pre-trained model for cat2dog on kaggle. Please let me know if you have any issues with it. I saved the results in this pdf so you can see what it looks like:
results.pdf I used the cat2dog dataset from DRIT.

It takes 4+ days to train cropped face dataset and 16+ days to train cropped body dataset on Nvidia GPUs (estimates). Since it takes many days to train the dataset once, and it takes many iterations of training it will take some time but eventually many people will publish and share their pre-trained models in the weeks to come. Datasets can be found at DRIT. For selfie2anime you can use datasets selfie and anime face dataset. Other potential anime face dataset sources: thiswaifudoesnotexist,
animeGAN for generating anime images and a one click download anime face dataset.
UGATIT is quite general, you really just need a folder of anime faces and a folder of human faces and it figures the rest by itself.

Thanks for your sharing!! Have you calculated the Flops of this UGATIT model?

[thewaifuai]

I have published a pre-trained model for cat2dog on kaggle. Please let me know if you have any issues with it. I saved the results in this pdf so you can see what it looks like:
results.pdf I used the cat2dog dataset from DRIT.
It takes 4+ days to train cropped face dataset and 16+ days to train cropped body dataset on Nvidia GPUs (estimates). Since it takes many days to train the dataset once, and it takes many iterations of training it will take some time but eventually many people will publish and share their pre-trained models in the weeks to come. Datasets can be found at DRIT. For selfie2anime you can use datasets selfie and anime face dataset. Other potential anime face dataset sources: thiswaifudoesnotexist,
animeGAN for generating anime images and a one click download anime face dataset.
UGATIT is quite general, you really just need a folder of anime faces and a folder of human faces and it figures the rest by itself.

Thanks for your sharing!! Have you calculated the Flops of this UGATIT model?

No, I have not.

[eddybogosian]

That's awesome. It seemed to do better going from dog to carry thrown than the other way around...

…

________________________________ From: WaifuAI <notifications@github.com> Sent: Friday, October 30, 2020 4:22:28 AM To: taki0112/UGATIT <UGATIT@noreply.github.com> Cc: eddybogosian <edbogosian@hotmail.com>; Manual <manual@noreply.github.com> Subject: Re: [taki0112/UGATIT] Pretrained model? (#5) I have published a pre-trained model for cat2dog on kaggle<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.kaggle.com%2Fwaifuai%2Fugatit-cat2dog-pretrained-model&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495478111%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=SPs6Ip4xo3a8txPVfOUjErzpn3G6kBmhKILk7fi0pSs%3D&reserved=0>. Please let me know if you have any issues with it. I saved the results in this pdf so you can see what it looks like: results.pdf<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ftaki0112%2FUGATIT%2Ffiles%2F3484431%2Fresults.pdf&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495488105%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=Zs5%2FgGZy%2BYjAZiOh0l9SH6j6cA49sB71x7%2BUoioPFmg%3D&reserved=0> I used the cat2dog dataset<https://nam11.safelinks.protection.outlook.com/?url=http%3A%2F%2Fvllab.ucmerced.edu%2Fhylee%2FDRIT%2F&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495488105%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=cR7bwyhyWoSwYyS%2FUp9o9kLRxg%2FleQfYY%2Ffvoih2OA4%3D&reserved=0> from DRIT<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ftaki0112%2FDRIT-Tensorflow%23download-dataset&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495488105%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=5rhtibj8jXnKy4vJLtV6LnSzCkrBTszjf5pZkxJoaGY%3D&reserved=0>. It takes 4+ days to train cropped face dataset and 16+ days to train cropped body dataset on Nvidia GPUs (estimates). Since it takes many days to train the dataset once, and it takes many iterations of training it will take some time but eventually many people will publish and share their pre-trained models in the weeks to come. Datasets can be found at DRIT<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ftaki0112%2FDRIT-Tensorflow%23download-dataset&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495498101%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=PfR1nwz2%2BZCkOfb%2BiU82JqkNE%2BbQ2tXYBT2PFqzHZ3o%3D&reserved=0>. For selfie2anime you can use datasets selfie<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.crcv.ucf.edu%2Fdata%2FSelfie%2F&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495498101%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=mJvd62vj22KS0u1gVlTouOmhIkFXki5zyyQWuV4b7V0%3D&reserved=0> and anime face dataset<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FMckinsey666%2FAnime-Face-Dataset&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495508093%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=hUdT2iNjMWF8HdQj4SHFsNbdSJAYa2X7lTFBxROTqZM%3D&reserved=0>. Other potential anime face dataset sources: thiswaifudoesnotexist<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.gwern.net%2FTWDNE%23downloads&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393495997811%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=GrDO4N%2FvMt8FNnj8OjRgu2bm3yA2UX7tpZEoByE0nF4%3D&reserved=0>, animeGAN<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fjayleicn%2FanimeGAN&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393496007798%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=zVfQr%2BNikWZv1swAG82WDt4e3F5NhkPV%2F9DHgYBhWlI%3D&reserved=0> for generating anime images and a one click download anime face dataset<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FMckinsey666%2FAnime-Face-Dataset&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393496007798%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=35K2xcsXYvvwQLgjOd0djZ7rqCWqJ%2B0crQLpnvHDYvk%3D&reserved=0>. UGATIT is quite general, you really just need a folder of anime faces and a folder of human faces and it figures the rest by itself. Thanks for your sharing!! Have you calculated the Flops of this UGATIT model? No, I have not. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ftaki0112%2FUGATIT%2Fissues%2F5%23issuecomment-719341251&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393496017793%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=hQCXSYKxz8cw%2Fxxm%2B3ETjK%2BEvxYBydohacix2hHA0Ks%3D&reserved=0>, or unsubscribe<https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FABZPKZJFRGE5PSF7DJDF6KDSNJSTJANCNFSM4II75NLA&data=04%7C01%7C%7C1b171a478e6040939be608d87ca48f18%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637396393496017793%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=firwWJyPGsfL8im1VjVa0FPkac3CeQXnK95hfgoG06o%3D&reserved=0>.