{
  "id": 98097,
  "title": "Building GANs to produce handwritten characters in TensorFlow",
  "url": "/competitions/generative-dog-images/discussion/98097",
  "author_name": "",
  "post_date": "2019-07-01T09:25:36.184203100Z",
  "votes": 4,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Some time ago I've written a chapter about GANs in the book <em>TensorFlow Deep Learning Projects</em> published by Packt (other co-authors were Alberto Boschetti, Alexey Grigorev, Rajalingappaa Shanmugamani, Abhishek Thakur who dealt with different chapters and topics in the book). </p>\n\n<p>The code has been written in TensorFlow and there are a few tricks for having GANs to converge faster that you may find useful, such as training the generator more times than the discriminator. You can find the code freely available here:</p>\n\n<p><a href=\"https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04\">https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04</a></p>\n\n<p>I hope you will find it useful (in the meantime I am preparing a public kernel that should work for this competition)</p>",
  "messages": [
    {
      "id": "565687",
      "postDate": "07/01/2019 09:25:36",
      "content": "<p>Some time ago I've written a chapter about GANs in the book <em>TensorFlow Deep Learning Projects</em> published by Packt (other co-authors were Alberto Boschetti, Alexey Grigorev, Rajalingappaa Shanmugamani, Abhishek Thakur who dealt with different chapters and topics in the book). </p>\n\n<p>The code has been written in TensorFlow and there are a few tricks for having GANs to converge faster that you may find useful, such as training the generator more times than the discriminator. You can find the code freely available here:</p>\n\n<p><a href=\"https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04\">https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04</a></p>\n\n<p>I hope you will find it useful (in the meantime I am preparing a public kernel that should work for this competition)</p>",
      "rawMarkdown": "Some time ago I've written a chapter about GANs in the book *TensorFlow Deep Learning Projects* published by Packt (other co-authors were Alberto Boschetti, Alexey Grigorev, Rajalingappaa Shanmugamani, Abhishek Thakur who dealt with different chapters and topics in the book). \n\nThe code has been written in TensorFlow and there are a few tricks for having GANs to converge faster that you may find useful, such as training the generator more times than the discriminator. You can find the code freely available here:\n\n[https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04](https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04)\n\nI hope you will find it useful (in the meantime I am preparing a public kernel that should work for this competition)",
      "votes": null
    },
    {
      "id": "565704",
      "postDate": "07/01/2019 09:42:44",
      "content": "<p>amazing material, thank you <a href=\"/lucamassaron\">@lucamassaron</a> 👍 😄 </p>",
      "rawMarkdown": "amazing material, thank you @lucamassaron 👍 😄",
      "votes": null
    },
    {
      "id": "566075",
      "postDate": "07/01/2019 19:12:03",
      "content": "<p>Thank you so much. I was fishing for GANs material , but they were all strict to mnist dataset. This is helpful !</p>",
      "rawMarkdown": "Thank you so much. I was fishing for GANs material , but they were all strict to mnist dataset. This is helpful !",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 565704,
      "author_name": "jesucristo",
      "author_url": "",
      "post_date": "07/01/2019 09:42:44",
      "content": "<p>amazing material, thank you <a href=\"/lucamassaron\">@lucamassaron</a> 👍 😄 </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 566075,
      "author_name": "rahul197",
      "author_url": "",
      "post_date": "07/01/2019 19:12:03",
      "content": "<p>Thank you so much. I was fishing for GANs material , but they were all strict to mnist dataset. This is helpful !</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "565687": "Some time ago I've written a chapter about GANs in the book *TensorFlow Deep Learning Projects* published by Packt (other co-authors were Alberto Boschetti, Alexey Grigorev, Rajalingappaa Shanmugamani, Abhishek Thakur who dealt with different chapters and topics in the book). \n\nThe code has been written in TensorFlow and there are a few tricks for having GANs to converge faster that you may find useful, such as training the generator more times than the discriminator. You can find the code freely available here:\n\n[https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04](https://github.com/lmassaron/TensorFlow-Deep-Learning-Projects/tree/master/Chapter04)\n\nI hope you will find it useful (in the meantime I am preparing a public kernel that should work for this competition)",
    "565704": "amazing material, thank you @lucamassaron 👍 😄",
    "566075": "Thank you so much. I was fishing for GANs material , but they were all strict to mnist dataset. This is helpful !"
  },
  "source": "meta"
}