{
  "id": 132538,
  "title": "custom dataset generator",
  "url": "/competitions/deepfake-detection-challenge/discussion/132538",
  "author_name": "rishabh12j",
  "post_date": "2020-02-26T14:08:51.557000",
  "votes": 0,
  "comment_count": 2,
  "views": 0,
  "content": "<p>can anyone please help to generate a custom data generator over data i have frames in 2 folders real and fake and i have to generate a data generator that takes next number of batch size images and create a data than again the next data and so on</p>",
  "messages": [
    {
      "id": 757190,
      "postDate": "2020-02-26T14:08:51.557Z",
      "content": "<p>can anyone please help to generate a custom data generator over data i have frames in 2 folders real and fake and i have to generate a data generator that takes next number of batch size images and create a data than again the next data and so on</p>",
      "rawMarkdown": "can anyone please help to generate a custom data generator over data i have frames in 2 folders real and fake and i have to generate a data generator that takes next number of batch size images and create a data than again the next data and so on"
    },
    {
      "id": 757277,
      "postDate": "2020-02-26T15:46:29.100Z",
      "content": "<p>For Pytorch :  You can Create the Dataset using torchvision <a href=\"https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder\">ImageFolder </a>API\nFor Keras/tf.Keras :  you can use <a href=\"https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator\">ImageDataGenerator </a>with flow from directory API </p>\n\n<p>When in doubt check starter kernels which are already available.</p>",
      "rawMarkdown": "For Pytorch :  You can Create the Dataset using torchvision [ImageFolder ](https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder)API\nFor Keras/tf.Keras :  you can use [ImageDataGenerator ](https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator)with flow from directory API \n\nWhen in doubt check starter kernels which are already available.",
      "isDeleted": true,
      "replies": [
        {
          "id": 761072,
          "postDate": "2020-03-02T05:05:24.503Z",
          "content": "<p>can i have a link of starter kernel also please</p>",
          "rawMarkdown": "can i have a link of starter kernel also please"
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 757277,
      "author_name": "",
      "author_url": "",
      "post_date": "2020-02-26T15:46:29.100000",
      "content": "<p>For Pytorch :  You can Create the Dataset using torchvision <a href=\"https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder\">ImageFolder </a>API\nFor Keras/tf.Keras :  you can use <a href=\"https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator\">ImageDataGenerator </a>with flow from directory API </p>\n\n<p>When in doubt check starter kernels which are already available.</p>",
      "votes": 0,
      "replies": [
        {
          "id": 761072,
          "author_name": "rishabh12j",
          "author_url": "",
          "post_date": "2020-03-02T05:05:24.503000",
          "content": "<p>can i have a link of starter kernel also please</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "757190": "can anyone please help to generate a custom data generator over data i have frames in 2 folders real and fake and i have to generate a data generator that takes next number of batch size images and create a data than again the next data and so on",
    "757277": "For Pytorch :  You can Create the Dataset using torchvision [ImageFolder ](https://pytorch.org/docs/stable/torchvision/datasets.html#imagefolder)API\nFor Keras/tf.Keras :  you can use [ImageDataGenerator ](https://www.tensorflow.org/api_docs/python/tf/keras/preprocessing/image/ImageDataGenerator)with flow from directory API \n\nWhen in doubt check starter kernels which are already available."
  }
}