{
  "id": 200748,
  "title": "What is Inference ?",
  "url": "/competitions/cassava-leaf-disease-classification/discussion/200748",
  "author_name": "",
  "post_date": "2020-12-01T17:20:00.511494800Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hello all, lately I am going through the kernels and I see <strong>inference</strong> everywhere  and I was wondering what <strong>inference</strong> actually meant?</p>",
  "messages": [
    {
      "id": "1098507",
      "postDate": "12/01/2020 17:20:00",
      "content": "<p>Hello all, lately I am going through the kernels and I see <strong>inference</strong> everywhere  and I was wondering what <strong>inference</strong> actually meant?</p>",
      "rawMarkdown": "Hello all, lately I am going through the kernels and I see **inference** everywhere  and I was wondering what **inference** actually meant?",
      "votes": null
    },
    {
      "id": "1098754",
      "postDate": "12/01/2020 20:23:04",
      "content": "<p>Inference is when you take your trained model and use it to predict the unknown, test dataset.</p>\n<p>Inference - root word Infer</p>\n<p>Because this competition requires Inference to be done in a committed notebook, we are referencing \"Inference\" more than usual. But it is the key step in every model - predicting the unknown data.</p>\n<p>-Rich</p>",
      "rawMarkdown": "Inference is when you take your trained model and use it to predict the unknown, test dataset.\n\nInference - root word Infer\n\nBecause this competition requires Inference to be done in a committed notebook, we are referencing \"Inference\" more than usual. But it is the key step in every model - predicting the unknown data.\n\n-Rich",
      "votes": null
    },
    {
      "id": "1099068",
      "postDate": "12/02/2020 04:16:59",
      "content": "<p>Thanks Rich, Now I finally understand the reason why everyone is loading an already trained model and using it for prediction.</p>",
      "rawMarkdown": "Thanks Rich, Now I finally understand the reason why everyone is loading an already trained model and using it for prediction.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1098754,
      "author_name": "richardepstein",
      "author_url": "",
      "post_date": "12/01/2020 20:23:04",
      "content": "<p>Inference is when you take your trained model and use it to predict the unknown, test dataset.</p>\n<p>Inference - root word Infer</p>\n<p>Because this competition requires Inference to be done in a committed notebook, we are referencing \"Inference\" more than usual. But it is the key step in every model - predicting the unknown data.</p>\n<p>-Rich</p>",
      "votes": null,
      "replies": [
        {
          "id": 1099068,
          "author_name": "asrithdasari",
          "author_url": "",
          "post_date": "12/02/2020 04:16:59",
          "content": "<p>Thanks Rich, Now I finally understand the reason why everyone is loading an already trained model and using it for prediction.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1098507": "Hello all, lately I am going through the kernels and I see **inference** everywhere  and I was wondering what **inference** actually meant?",
    "1098754": "Inference is when you take your trained model and use it to predict the unknown, test dataset.\n\nInference - root word Infer\n\nBecause this competition requires Inference to be done in a committed notebook, we are referencing \"Inference\" more than usual. But it is the key step in every model - predicting the unknown data.\n\n-Rich",
    "1099068": "Thanks Rich, Now I finally understand the reason why everyone is loading an already trained model and using it for prediction."
  },
  "source": "meta"
}