{
  "id": 185141,
  "title": "How did they get 1.000 score?",
  "url": "/competitions/digit-recognizer/discussion/185141",
  "author_name": "绿箭薄荷糖",
  "post_date": "2020-09-19T14:02:55.579000",
  "votes": 2,
  "comment_count": 2,
  "views": null,
  "content": "<p>Why does many people can reach 1.0000 scores? That is amazing.</p>",
  "messages": [
    {
      "id": 1018287,
      "postDate": "2020-09-19T15:24:01.107Z",
      "content": "<p>I scored 1.00 because I figured out that the images present in the train and test set were directly taken from the labelled mnist data from keras datasets and using that one can easily predict without making any model.</p>\n<p>The maximum score one could achieve using a cnn model would be about 99.8-99.9.</p>\n<p>Please note that in no way I am promoting this way of participating in the competition. That's why I created and trained a cnn model. </p>",
      "rawMarkdown": "I scored 1.00 because I figured out that the images present in the train and test set were directly taken from the labelled mnist data from keras datasets and using that one can easily predict without making any model.\n\nThe maximum score one could achieve using a cnn model would be about 99.8-99.9.\n\nPlease note that in no way I am promoting this way of participating in the competition. That's why I created and trained a cnn model. ",
      "votes": 3
    },
    {
      "id": 1018197,
      "postDate": "2020-09-19T14:02:55.580Z",
      "content": "<p>Why does many people can reach 1.0000 scores? That is amazing.</p>",
      "rawMarkdown": "Why does many people can reach 1.0000 scores? That is amazing.",
      "votes": 2
    },
    {
      "id": 1028718,
      "postDate": "2020-09-27T06:20:45.610Z",
      "content": "<p>Quite simple. Here is an example: <a href=\"https://www.kaggle.com/bgmello/how-top-lb-got-their-score-use-mnist-to-learn\" target=\"_blank\">https://www.kaggle.com/bgmello/how-top-lb-got-their-score-use-mnist-to-learn</a></p>",
      "rawMarkdown": "Quite simple. Here is an example: https://www.kaggle.com/bgmello/how-top-lb-got-their-score-use-mnist-to-learn",
      "votes": 1
    }
  ],
  "comments": [
    {
      "id": 1018287,
      "author_name": "Parth Chhabra",
      "author_url": "",
      "post_date": "2020-09-19T15:24:01.107000",
      "content": "<p>I scored 1.00 because I figured out that the images present in the train and test set were directly taken from the labelled mnist data from keras datasets and using that one can easily predict without making any model.</p>\n<p>The maximum score one could achieve using a cnn model would be about 99.8-99.9.</p>\n<p>Please note that in no way I am promoting this way of participating in the competition. That's why I created and trained a cnn model. </p>",
      "votes": 3,
      "replies": []
    },
    {
      "id": 1028718,
      "author_name": "Alan Choon",
      "author_url": "",
      "post_date": "2020-09-27T06:20:45.610000",
      "content": "<p>Quite simple. Here is an example: <a href=\"https://www.kaggle.com/bgmello/how-top-lb-got-their-score-use-mnist-to-learn\" target=\"_blank\">https://www.kaggle.com/bgmello/how-top-lb-got-their-score-use-mnist-to-learn</a></p>",
      "votes": 1,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1018287": "I scored 1.00 because I figured out that the images present in the train and test set were directly taken from the labelled mnist data from keras datasets and using that one can easily predict without making any model.\n\nThe maximum score one could achieve using a cnn model would be about 99.8-99.9.\n\nPlease note that in no way I am promoting this way of participating in the competition. That's why I created and trained a cnn model. ",
    "1018197": "Why does many people can reach 1.0000 scores? That is amazing.",
    "1028718": "Quite simple. Here is an example: https://www.kaggle.com/bgmello/how-top-lb-got-their-score-use-mnist-to-learn"
  }
}