{
  "id": 241461,
  "title": "what is the photo_jpg folder for?",
  "url": "/competitions/gan-getting-started/discussion/241461",
  "author_name": "",
  "post_date": "2021-05-24T16:18:06.385162700Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>our initial guess:<br>\nfeeding the photo_… folder data into the GAN to transform them into the results.<br>\nis that correct?</p>",
  "messages": [
    {
      "id": "1321338",
      "postDate": "05/24/2021 16:18:06",
      "content": "<p>our initial guess:<br>\nfeeding the photo_… folder data into the GAN to transform them into the results.<br>\nis that correct?</p>",
      "rawMarkdown": "our initial guess:\nfeeding the photo_... folder data into the GAN to transform them into the results.\nis that correct?",
      "votes": null
    },
    {
      "id": "1355347",
      "postDate": "06/18/2021 08:41:28",
      "content": "<p>I think it is there as an alternative to the photo_tfrec folder, the dataset is the same but in different format if you want to use different training methods and dont want to use tfrec format</p>",
      "rawMarkdown": "I think it is there as an alternative to the photo_tfrec folder, the dataset is the same but in different format if you want to use different training methods and dont want to use tfrec format",
      "votes": null
    },
    {
      "id": "1492524",
      "postDate": "08/27/2021 08:29:28",
      "content": "<p>That is the pictures you want to augment by pushing it through the Generator so that the Discriminator believes that Monet painted it. The Discriminator gets two inputs a real and a fake image. The Generator job is to change its input so that it fools the Discriminator. </p>\n<p>When you have a satisfying model then you can push random pictures through the generator to make it more Monet-like. So yes, you are correct. <br>\nFor a bit more context: <br>\n<a href=\"https://wiki.pathmind.com/generative-adversarial-network-gan\" target=\"_blank\">https://wiki.pathmind.com/generative-adversarial-network-gan</a></p>",
      "rawMarkdown": "That is the pictures you want to augment by pushing it through the Generator so that the Discriminator believes that Monet painted it. The Discriminator gets two inputs a real and a fake image. The Generator job is to change its input so that it fools the Discriminator. \n\nWhen you have a satisfying model then you can push random pictures through the generator to make it more Monet-like. So yes, you are correct. \nFor a bit more context: \nhttps://wiki.pathmind.com/generative-adversarial-network-gan",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1355347,
      "author_name": "kartikchoudhary727",
      "author_url": "",
      "post_date": "06/18/2021 08:41:28",
      "content": "<p>I think it is there as an alternative to the photo_tfrec folder, the dataset is the same but in different format if you want to use different training methods and dont want to use tfrec format</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1492524,
      "author_name": "emilskaaning",
      "author_url": "",
      "post_date": "08/27/2021 08:29:28",
      "content": "<p>That is the pictures you want to augment by pushing it through the Generator so that the Discriminator believes that Monet painted it. The Discriminator gets two inputs a real and a fake image. The Generator job is to change its input so that it fools the Discriminator. </p>\n<p>When you have a satisfying model then you can push random pictures through the generator to make it more Monet-like. So yes, you are correct. <br>\nFor a bit more context: <br>\n<a href=\"https://wiki.pathmind.com/generative-adversarial-network-gan\" target=\"_blank\">https://wiki.pathmind.com/generative-adversarial-network-gan</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1321338": "our initial guess:\nfeeding the photo_... folder data into the GAN to transform them into the results.\nis that correct?",
    "1355347": "I think it is there as an alternative to the photo_tfrec folder, the dataset is the same but in different format if you want to use different training methods and dont want to use tfrec format",
    "1492524": "That is the pictures you want to augment by pushing it through the Generator so that the Discriminator believes that Monet painted it. The Discriminator gets two inputs a real and a fake image. The Generator job is to change its input so that it fools the Discriminator. \n\nWhen you have a satisfying model then you can push random pictures through the generator to make it more Monet-like. So yes, you are correct. \nFor a bit more context: \nhttps://wiki.pathmind.com/generative-adversarial-network-gan"
  },
  "source": "meta"
}