{
  "id": 484074,
  "title": "Can this dataset be used for training DDPM？",
  "url": "/competitions/dog-breed-identification/discussion/484074",
  "author_name": "SukiLiyuu_",
  "post_date": "2024-03-15T08:54:41.714000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Recently, I've been wanting to use a diffusion model like DDPM to generate pet images, and after a failed attempt with the Oxford-IIIT Pet dataset, I think one reason for the failure might be the small size of the training set. So, can this dataset be used for DDPM, and has anyone tried this?</p>",
  "messages": [
    {
      "id": 2698041,
      "postDate": "2024-03-15T08:54:41.713Z",
      "content": "<p>Recently, I've been wanting to use a diffusion model like DDPM to generate pet images, and after a failed attempt with the Oxford-IIIT Pet dataset, I think one reason for the failure might be the small size of the training set. So, can this dataset be used for DDPM, and has anyone tried this?</p>",
      "rawMarkdown": "Recently, I've been wanting to use a diffusion model like DDPM to generate pet images, and after a failed attempt with the Oxford-IIIT Pet dataset, I think one reason for the failure might be the small size of the training set. So, can this dataset be used for DDPM, and has anyone tried this?"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2698041": "Recently, I've been wanting to use a diffusion model like DDPM to generate pet images, and after a failed attempt with the Oxford-IIIT Pet dataset, I think one reason for the failure might be the small size of the training set. So, can this dataset be used for DDPM, and has anyone tried this?"
  }
}