{
  "id": 50916,
  "title": "pytorch with only the pretrained models from imagenet",
  "url": "/competitions/dog-breed-identification/discussion/50916",
  "author_name": "",
  "post_date": "2018-03-02T17:29:26.364380900Z",
  "votes": 6,
  "comment_count": 2,
  "views": 0,
  "content": "<p><a href=\"https://github.com/TreB1eN/Dog-breed-competition-with-only-the-pretrained-models-from-imagenet\">https://github.com/TreB1eN/Dog-breed-competition-with-only-the-pretrained-models-from-imagenet</a></p>\n\n<p>This post describe the method and tips I got from participanting in the Dog Breed challenge[Dog Breed challenge] in Kaggle. I managed to get an final score of 0.13783, which sets me in the 158 postion, I believe my approach worth sharing cause there is nothing else being used except the pretrained imagenet models.</p>",
  "messages": [
    {
      "id": "289971",
      "postDate": "03/02/2018 17:29:26",
      "content": "<p><a href=\"https://github.com/TreB1eN/Dog-breed-competition-with-only-the-pretrained-models-from-imagenet\">https://github.com/TreB1eN/Dog-breed-competition-with-only-the-pretrained-models-from-imagenet</a></p>\n\n<p>This post describe the method and tips I got from participanting in the Dog Breed challenge[Dog Breed challenge] in Kaggle. I managed to get an final score of 0.13783, which sets me in the 158 postion, I believe my approach worth sharing cause there is nothing else being used except the pretrained imagenet models.</p>",
      "rawMarkdown": "https://github.com/TreB1eN/Dog-breed-competition-with-only-the-pretrained-models-from-imagenet\n\nThis post describe the method and tips I got from participanting in the Dog Breed challenge[Dog Breed challenge] in Kaggle. I managed to get an final score of 0.13783, which sets me in the 158 postion, I believe my approach worth sharing cause there is nothing else being used except the pretrained imagenet models.",
      "votes": null
    },
    {
      "id": "289972",
      "postDate": "03/02/2018 17:32:49",
      "content": "<p>Good job man. My suggestion is that you can turn this into a Python notebook for better readability!</p>",
      "rawMarkdown": "Good job man. My suggestion is that you can turn this into a Python notebook for better readability!",
      "votes": null
    },
    {
      "id": "289973",
      "postDate": "03/02/2018 17:43:33",
      "content": "<p>Thanks, already done</p>",
      "rawMarkdown": "Thanks, already done",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 289972,
      "author_name": "pukkinming",
      "author_url": "",
      "post_date": "03/02/2018 17:32:49",
      "content": "<p>Good job man. My suggestion is that you can turn this into a Python notebook for better readability!</p>",
      "votes": null,
      "replies": [
        {
          "id": 289973,
          "author_name": "",
          "author_url": "",
          "post_date": "03/02/2018 17:43:33",
          "content": "<p>Thanks, already done</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "289971": "https://github.com/TreB1eN/Dog-breed-competition-with-only-the-pretrained-models-from-imagenet\n\nThis post describe the method and tips I got from participanting in the Dog Breed challenge[Dog Breed challenge] in Kaggle. I managed to get an final score of 0.13783, which sets me in the 158 postion, I believe my approach worth sharing cause there is nothing else being used except the pretrained imagenet models.",
    "289972": "Good job man. My suggestion is that you can turn this into a Python notebook for better readability!",
    "289973": "Thanks, already done"
  },
  "source": "meta"
}