{ "id": 178253, "title": "Complete GAN Resources (so far)", "url": "/competitions/gan-getting-started/discussion/178253", "author_name": "", "post_date": "2020-08-29T08:42:23.504709Z", "votes": 24, "comment_count": 3, "views": 0, "content": "

Here are listed out some GAN resources:

\n
\n

Book

\n\n

The most book covers the following GAN cases:

\n
- Standard: GAN, DCGAN.\n- Conditional: cGAN, SS-GAN, InfoGAN, ACGAN.\n- Loss: WGAN, WGAN-GP, LSGAN.\n- Image Translation: Pix2Pix, CycleGAN.\n- Advanced GANs: BigGAN, PG-GAN, StyleGAN.\n- Other: StackGAN, 3DGAN, BEGAN, SRGAN, DiscoGAN, SEGAN\n
\n
\n

Codes

\n

Using PyTorch, here are different type of GAN architecture implemented using PyTorch, namely:

\n\n

More GAN-codes using PyTorch, PyTorch-GAN:

\n\n

Using Keras, [Keras-GAN](https://github.com/eriklindernoren/Keras-GAN, namely:

\n\n
\n

Videos

\n\n
\n

Kaggle

\n

Kaggle, Previous Competition: Generative Dog Images. Here are the top 5 solutions:

\n\n
\n

Blog

\n

Jason Brownlee provides a numerous number of GAN tutorials. Check here. Here are some Jason tips to stabilize GAN, quoted from here.

\n
- Downsample Using Stride Convolutions\n- Upsample Using Stride Convolutions\n- Use LeakyReLU\n- Use Batch Normalization\n- Use Gaussian Weight Initialization\n- Use Adam Stochastic Gradient Descent\n- Scale Images to the Range [-1,1]\n- Use a Gaussian Latent Space\n- Separate Batches of Real and Fake Images\n- Use Label Smoothing\n- Use Noisy Labels\n
\n

Original sources: How to Train a GAN? Tips and tricks to make GANs work, ganhacks, PyTorch

\n
\n

GAN Applications

\n

Here is another great blog, discussing the application of GAN by Jonathan Hui - GAN — Some cool applications of GAN, namely:

\n
- Create Anime characters\n- Pose Guided Person Image Generation\n- CycleGAN\n- StarGAN\n- PixelDTGAN\n- Super-resolution\n- The progressive growing of GANs\n- StyleGAN2\n- High-resolution image synthesis\n- GauGAN\n- Text to Image Synthesis\n- Face synthesis\n- Image inpainting\n- Learn Joint Distribution\n- DiscoGAN\n- Pix2Pix\n- DTN\n- Texture synthesis\n- Image editing (IcGAN)\n- Face aging (Age-cGAN)\n- DeblurGAN\n- Neural Photo Editor\n- Object detection\n- Image blending\n- Video generation\n- Generate 3D objects\n- Music generation\n- Medical (Anomaly Detection)\n
\n
\n

GAN Playground

\n\n

\"UI\"

", "messages": [ { "id": "989971", "postDate": "08/29/2020 08:42:23", "content": "

Here are listed out some GAN resources:

\n
\n

Book

\n\n

The most book covers the following GAN cases:

\n
- Standard: GAN, DCGAN.\n- Conditional: cGAN, SS-GAN, InfoGAN, ACGAN.\n- Loss: WGAN, WGAN-GP, LSGAN.\n- Image Translation: Pix2Pix, CycleGAN.\n- Advanced GANs: BigGAN, PG-GAN, StyleGAN.\n- Other: StackGAN, 3DGAN, BEGAN, SRGAN, DiscoGAN, SEGAN\n
\n
\n

Codes

\n

Using PyTorch, here are different type of GAN architecture implemented using PyTorch, namely:

\n\n

More GAN-codes using PyTorch, PyTorch-GAN:

\n\n

Using Keras, [Keras-GAN](https://github.com/eriklindernoren/Keras-GAN, namely:

\n\n
\n

Videos

\n\n
\n

Kaggle

\n

Kaggle, Previous Competition: Generative Dog Images. Here are the top 5 solutions:

\n\n
\n

Blog

\n

Jason Brownlee provides a numerous number of GAN tutorials. Check here. Here are some Jason tips to stabilize GAN, quoted from here.

\n
- Downsample Using Stride Convolutions\n- Upsample Using Stride Convolutions\n- Use LeakyReLU\n- Use Batch Normalization\n- Use Gaussian Weight Initialization\n- Use Adam Stochastic Gradient Descent\n- Scale Images to the Range [-1,1]\n- Use a Gaussian Latent Space\n- Separate Batches of Real and Fake Images\n- Use Label Smoothing\n- Use Noisy Labels\n
\n

Original sources: How to Train a GAN? Tips and tricks to make GANs work, ganhacks, PyTorch

\n
\n

GAN Applications

\n

Here is another great blog, discussing the application of GAN by Jonathan Hui - GAN — Some cool applications of GAN, namely:

\n
- Create Anime characters\n- Pose Guided Person Image Generation\n- CycleGAN\n- StarGAN\n- PixelDTGAN\n- Super-resolution\n- The progressive growing of GANs\n- StyleGAN2\n- High-resolution image synthesis\n- GauGAN\n- Text to Image Synthesis\n- Face synthesis\n- Image inpainting\n- Learn Joint Distribution\n- DiscoGAN\n- Pix2Pix\n- DTN\n- Texture synthesis\n- Image editing (IcGAN)\n- Face aging (Age-cGAN)\n- DeblurGAN\n- Neural Photo Editor\n- Object detection\n- Image blending\n- Video generation\n- Generate 3D objects\n- Music generation\n- Medical (Anomaly Detection)\n
\n
\n

GAN Playground

\n\n

\"UI\"

", "rawMarkdown": "Here are listed out some GAN resources:\n\n---\n\n# Book\n\n- [Chapter 20. Deep Generative Models, Deep Learning, 2016](https://www.deeplearningbook.org/contents/generative_models.html) by [Ian Goodfellow](https://www.linkedin.com/in/ian-goodfellow-b7187213/)\n\n- [Chapter 8. Generative Deep Learning, Deep Learning with Python, 2017](https://www.amazon.com/Deep-Learning-Python-Francois-Chollet/dp/1617294438/ref=as_li_ss_tl?keywords=deep+learning&qid=1553730173&s=books&sr=1-2&linkCode=sl1&tag=inspiredalgor-20&linkId=680904778b3cf1d6c567f3ff3ec4b48b&language=en_US) by [François Chollet](https://www.linkedin.com/in/fchollet/)\n\n- [ Generative Deep Learning](https://www.amazon.com/Generative-Deep-Learning-Teaching-Machines/dp/1492041947/ref=as_li_ss_tl?keywords=generative+adversarial+networks&qid=1553729660&s=books&sr=1-9&linkCode=sl1&tag=inspiredalgor-20&linkId=76c68cafd5cea9370084a635dd025215&language=en_US)\n\n- [GANs in Action](https://www.amazon.com/GANs-Action-learning-Generative-Adversarial/dp/1617295566/ref=as_li_ss_tl?keywords=generative+adversarial+networks&qid=1553729660&s=books&sr=1-4&linkCode=sl1&tag=inspiredalgor-20&linkId=5d3236c158da6eaf60a544b58fbadc14&language=en_US) - [Code](https://github.com/GANs-in-Action/gans-in-action)\n\nThe most book covers the following GAN cases:\n\n```\n- Standard: GAN, DCGAN.\n- Conditional: cGAN, SS-GAN, InfoGAN, ACGAN.\n- Loss: WGAN, WGAN-GP, LSGAN.\n- Image Translation: Pix2Pix, CycleGAN.\n- Advanced GANs: BigGAN, PG-GAN, StyleGAN.\n- Other: StackGAN, 3DGAN, BEGAN, SRGAN, DiscoGAN, SEGAN\n```\n\n---\n\n# Codes\n\nUsing **PyTorch**, here are different type of [GAN architecture implemented using PyTorch](https://github.com/ozanciga/gans-with-pytorch), namely: \n\n- [Conditional Generative Adversarial Nets](https://github.com/ozanciga/gans-with-pytorch#cgan)\n- [CT-GAN: Wasserstein GANs: A Consistency Term and Its Dual Effect](https://github.com/ozanciga/gans-with-pytorch#ct-gan)\n- [InfoGAN](https://github.com/ozanciga/gans-with-pytorch#infogan)\n- [pix2pix](https://github.com/ozanciga/gans-with-pytorch#pix2pix)\n- [SRGAN](https://github.com/ozanciga/gans-with-pytorch#srgan)\n\nMore GAN-codes using **PyTorch**, [PyTorch-GAN](https://github.com/eriklindernoren/PyTorch-GAN):\n\n- [CycleGAN](https://github.com/eriklindernoren/PyTorch-GAN#cyclegan)\n- [CycleGAN](https://github.com/aitorzip/PyTorch-CycleGAN)\n- [Pix2Pix](https://github.com/eriklindernoren/PyTorch-GAN#pix2pix)\n\n\nUsing **Keras**, [Keras-GAN](https://github.com/eriklindernoren/Keras-GAN, namely: \n\n- [CycleGAN](https://github.com/eriklindernoren/Keras-GAN#cyclegan)\n- [Pix2Pix](https://github.com/eriklindernoren/Keras-GAN#pix2pix)\n- More: Auxiliary Classifier GAN, Adversarial Autoencoder, Bidirectional GAN, Boundary-Seeking GAN, Conditional GAN, Context-Conditional GAN, Context Encoder, Coupled GANs, Deep Convolutional GAN, DiscoGAN, DualGAN, InfoGAN, LSGAN, PixelDA, Semi-Supervised, Super-Resolution GAN, Wasserstein GAN, Wasserstein GAN GP\n\n---\n\n# Videos\n\n- [Generative Adversarial Networks](https://www.youtube.com/watch?v=HN9NRhm9waY) by [Ian Goodfellow](https://www.linkedin.com/in/ian-goodfellow-b7187213/)\n- [Adversarial Machine Learning: GAN](https://www.youtube.com/watch?v=AJJRWFVfNPg) by [Ian Goodfellow](https://www.linkedin.com/in/ian-goodfellow-b7187213/)\n- [Generative Models](https://www.youtube.com/watch?v=5WoItGTWV54) by Stanford University School of Engineering\n\n---\n\n# Kaggle \n\nKaggle, Previous Competition: [Generative Dog Images](https://www.kaggle.com/c/generative-dog-images/overview). Here are the top 5 solutions:\n\n- [1st Place](https://www.kaggle.com/c/generative-dog-images/discussion/106324) \n- [2nd Place](https://www.kaggle.com/c/generative-dog-images/discussion/104211)\n- [3rd Place](https://www.kaggle.com/c/generative-dog-images/discussion/106514)\n- [4th Place](https://www.kaggle.com/c/generative-dog-images/discussion/104281)\n- [5th Place](https://www.kaggle.com/c/generative-dog-images/discussion/104287)\n\n---\n\n# Blog\n\n[Jason Brownlee](https://machinelearningmastery.com/) provides a numerous number of GAN tutorials. Check [here](https://machinelearningmastery.com/category/generative-adversarial-networks/). Here are some [Jason tips](https://machinelearningmastery.com/how-to-train-stable-generative-adversarial-networks/) to stabilize GAN, quoted from [here](https://www.kaggle.com/c/generative-dog-images/discussion/98595).\n\n```\n- Downsample Using Stride Convolutions\n- Upsample Using Stride Convolutions\n- Use LeakyReLU\n- Use Batch Normalization\n- Use Gaussian Weight Initialization\n- Use Adam Stochastic Gradient Descent\n- Scale Images to the Range [-1,1]\n- Use a Gaussian Latent Space\n- Separate Batches of Real and Fake Images\n- Use Label Smoothing\n- Use Noisy Labels\n```\n\nOriginal sources: [How to Train a GAN? Tips and tricks to make GANs work, ganhacks, PyTorch](https://github.com/soumith/ganhacks)\n\n---\n\n# GAN Applications\n\nHere is another great blog, discussing the application of GAN by [Jonathan Hui](https://medium.com/@jonathan_hui) - [GAN — Some cool applications of GAN](https://medium.com/@jonathan_hui/gan-some-cool-applications-of-gans-4c9ecca35900), namely:\n\n```\n- Create Anime characters\n- Pose Guided Person Image Generation\n- CycleGAN\n- StarGAN\n- PixelDTGAN\n- Super-resolution\n- The progressive growing of GANs\n- StyleGAN2\n- High-resolution image synthesis\n- GauGAN\n- Text to Image Synthesis\n- Face synthesis\n- Image inpainting\n- Learn Joint Distribution\n- DiscoGAN\n- Pix2Pix\n- DTN\n- Texture synthesis\n- Image editing (IcGAN)\n- Face aging (Age-cGAN)\n- DeblurGAN\n- Neural Photo Editor\n- Object detection\n- Image blending\n- Video generation\n- Generate 3D objects\n- Music generation\n- Medical (Anomaly Detection)\n```\n\n---\n\n# GAN Playground\n\n- [GAN Lab](https://poloclub.github.io/ganlab/)\n\n![UI](https://user-images.githubusercontent.com/17668390/91633072-2edb0e00-ea07-11ea-8512-5f4ba3b5365d.gif)", "votes": null }, { "id": "989985", "postDate": "08/29/2020 08:52:19", "content": "

thanks for share

", "rawMarkdown": "thanks for share", "votes": null }, { "id": "990018", "postDate": "08/29/2020 09:22:23", "content": "

Nice information!!
\nThanks for sharing.

", "rawMarkdown": "Nice information!!\nThanks for sharing.", "votes": null }, { "id": "2111973", "postDate": "01/23/2023 10:26:37", "content": "

This is a great source! Thank you

", "rawMarkdown": "This is a great source! Thank you", "votes": null } ], "comments": [ { "id": 989985, "author_name": "singhakash", "author_url": "", "post_date": "08/29/2020 08:52:19", "content": "

thanks for share

", "votes": null, "replies": [] }, { "id": 990018, "author_name": "akshaychavan123", "author_url": "", "post_date": "08/29/2020 09:22:23", "content": "

Nice information!!
\nThanks for sharing.

", "votes": null, "replies": [] }, { "id": 2111973, "author_name": "bell2psy", "author_url": "", "post_date": "01/23/2023 10:26:37", "content": "

This is a great source! Thank you

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "989971": "Here are listed out some GAN resources:\n\n---\n\n# Book\n\n- [Chapter 20. Deep Generative Models, Deep Learning, 2016](https://www.deeplearningbook.org/contents/generative_models.html) by [Ian Goodfellow](https://www.linkedin.com/in/ian-goodfellow-b7187213/)\n\n- [Chapter 8. Generative Deep Learning, Deep Learning with Python, 2017](https://www.amazon.com/Deep-Learning-Python-Francois-Chollet/dp/1617294438/ref=as_li_ss_tl?keywords=deep+learning&qid=1553730173&s=books&sr=1-2&linkCode=sl1&tag=inspiredalgor-20&linkId=680904778b3cf1d6c567f3ff3ec4b48b&language=en_US) by [François Chollet](https://www.linkedin.com/in/fchollet/)\n\n- [ Generative Deep Learning](https://www.amazon.com/Generative-Deep-Learning-Teaching-Machines/dp/1492041947/ref=as_li_ss_tl?keywords=generative+adversarial+networks&qid=1553729660&s=books&sr=1-9&linkCode=sl1&tag=inspiredalgor-20&linkId=76c68cafd5cea9370084a635dd025215&language=en_US)\n\n- [GANs in Action](https://www.amazon.com/GANs-Action-learning-Generative-Adversarial/dp/1617295566/ref=as_li_ss_tl?keywords=generative+adversarial+networks&qid=1553729660&s=books&sr=1-4&linkCode=sl1&tag=inspiredalgor-20&linkId=5d3236c158da6eaf60a544b58fbadc14&language=en_US) - [Code](https://github.com/GANs-in-Action/gans-in-action)\n\nThe most book covers the following GAN cases:\n\n```\n- Standard: GAN, DCGAN.\n- Conditional: cGAN, SS-GAN, InfoGAN, ACGAN.\n- Loss: WGAN, WGAN-GP, LSGAN.\n- Image Translation: Pix2Pix, CycleGAN.\n- Advanced GANs: BigGAN, PG-GAN, StyleGAN.\n- Other: StackGAN, 3DGAN, BEGAN, SRGAN, DiscoGAN, SEGAN\n```\n\n---\n\n# Codes\n\nUsing **PyTorch**, here are different type of [GAN architecture implemented using PyTorch](https://github.com/ozanciga/gans-with-pytorch), namely: \n\n- [Conditional Generative Adversarial Nets](https://github.com/ozanciga/gans-with-pytorch#cgan)\n- [CT-GAN: Wasserstein GANs: A Consistency Term and Its Dual Effect](https://github.com/ozanciga/gans-with-pytorch#ct-gan)\n- [InfoGAN](https://github.com/ozanciga/gans-with-pytorch#infogan)\n- [pix2pix](https://github.com/ozanciga/gans-with-pytorch#pix2pix)\n- [SRGAN](https://github.com/ozanciga/gans-with-pytorch#srgan)\n\nMore GAN-codes using **PyTorch**, [PyTorch-GAN](https://github.com/eriklindernoren/PyTorch-GAN):\n\n- [CycleGAN](https://github.com/eriklindernoren/PyTorch-GAN#cyclegan)\n- [CycleGAN](https://github.com/aitorzip/PyTorch-CycleGAN)\n- [Pix2Pix](https://github.com/eriklindernoren/PyTorch-GAN#pix2pix)\n\n\nUsing **Keras**, [Keras-GAN](https://github.com/eriklindernoren/Keras-GAN, namely: \n\n- [CycleGAN](https://github.com/eriklindernoren/Keras-GAN#cyclegan)\n- [Pix2Pix](https://github.com/eriklindernoren/Keras-GAN#pix2pix)\n- More: Auxiliary Classifier GAN, Adversarial Autoencoder, Bidirectional GAN, Boundary-Seeking GAN, Conditional GAN, Context-Conditional GAN, Context Encoder, Coupled GANs, Deep Convolutional GAN, DiscoGAN, DualGAN, InfoGAN, LSGAN, PixelDA, Semi-Supervised, Super-Resolution GAN, Wasserstein GAN, Wasserstein GAN GP\n\n---\n\n# Videos\n\n- [Generative Adversarial Networks](https://www.youtube.com/watch?v=HN9NRhm9waY) by [Ian Goodfellow](https://www.linkedin.com/in/ian-goodfellow-b7187213/)\n- [Adversarial Machine Learning: GAN](https://www.youtube.com/watch?v=AJJRWFVfNPg) by [Ian Goodfellow](https://www.linkedin.com/in/ian-goodfellow-b7187213/)\n- [Generative Models](https://www.youtube.com/watch?v=5WoItGTWV54) by Stanford University School of Engineering\n\n---\n\n# Kaggle \n\nKaggle, Previous Competition: [Generative Dog Images](https://www.kaggle.com/c/generative-dog-images/overview). Here are the top 5 solutions:\n\n- [1st Place](https://www.kaggle.com/c/generative-dog-images/discussion/106324) \n- [2nd Place](https://www.kaggle.com/c/generative-dog-images/discussion/104211)\n- [3rd Place](https://www.kaggle.com/c/generative-dog-images/discussion/106514)\n- [4th Place](https://www.kaggle.com/c/generative-dog-images/discussion/104281)\n- [5th Place](https://www.kaggle.com/c/generative-dog-images/discussion/104287)\n\n---\n\n# Blog\n\n[Jason Brownlee](https://machinelearningmastery.com/) provides a numerous number of GAN tutorials. Check [here](https://machinelearningmastery.com/category/generative-adversarial-networks/). Here are some [Jason tips](https://machinelearningmastery.com/how-to-train-stable-generative-adversarial-networks/) to stabilize GAN, quoted from [here](https://www.kaggle.com/c/generative-dog-images/discussion/98595).\n\n```\n- Downsample Using Stride Convolutions\n- Upsample Using Stride Convolutions\n- Use LeakyReLU\n- Use Batch Normalization\n- Use Gaussian Weight Initialization\n- Use Adam Stochastic Gradient Descent\n- Scale Images to the Range [-1,1]\n- Use a Gaussian Latent Space\n- Separate Batches of Real and Fake Images\n- Use Label Smoothing\n- Use Noisy Labels\n```\n\nOriginal sources: [How to Train a GAN? Tips and tricks to make GANs work, ganhacks, PyTorch](https://github.com/soumith/ganhacks)\n\n---\n\n# GAN Applications\n\nHere is another great blog, discussing the application of GAN by [Jonathan Hui](https://medium.com/@jonathan_hui) - [GAN — Some cool applications of GAN](https://medium.com/@jonathan_hui/gan-some-cool-applications-of-gans-4c9ecca35900), namely:\n\n```\n- Create Anime characters\n- Pose Guided Person Image Generation\n- CycleGAN\n- StarGAN\n- PixelDTGAN\n- Super-resolution\n- The progressive growing of GANs\n- StyleGAN2\n- High-resolution image synthesis\n- GauGAN\n- Text to Image Synthesis\n- Face synthesis\n- Image inpainting\n- Learn Joint Distribution\n- DiscoGAN\n- Pix2Pix\n- DTN\n- Texture synthesis\n- Image editing (IcGAN)\n- Face aging (Age-cGAN)\n- DeblurGAN\n- Neural Photo Editor\n- Object detection\n- Image blending\n- Video generation\n- Generate 3D objects\n- Music generation\n- Medical (Anomaly Detection)\n```\n\n---\n\n# GAN Playground\n\n- [GAN Lab](https://poloclub.github.io/ganlab/)\n\n![UI](https://user-images.githubusercontent.com/17668390/91633072-2edb0e00-ea07-11ea-8512-5f4ba3b5365d.gif)", "989985": "thanks for share", "990018": "Nice information!!\nThanks for sharing.", "2111973": "This is a great source! Thank you" }, "source": "meta" }