{
  "id": 491046,
  "title": "Calculate WEEK_NUM & MONTH by date_decision in test data?",
  "url": "/competitions/home-credit-credit-risk-model-stability/discussion/491046",
  "author_name": "",
  "post_date": "2024-04-04T12:09:47.078608900Z",
  "votes": 1,
  "comment_count": 4,
  "views": 0,
  "content": "<p>Hi, I am not sure if I understand it correctly, but isn't that the fact that MONTH and WEEK_NUM columns can be directly calculated from date_decision?<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2Ff9916bef40ff638c6d0322a90c569e4b%2F2024-04-04%20144508.png?generation=1712213197463571&amp;alt=media\" alt=\"MONTH\"><br>\nif so, what is the meaning of setting WEEK_NUM and MONTH as constant in the given test data?<br>\nWhat's more, do the date_decision corresponds to real date when the decision is made? I am thinking about important relevant US interest rate and other data to facilitate my modeling?<br>\nCan anyone help me out? many thanks.</p>",
  "messages": [
    {
      "id": "2734930",
      "postDate": "04/04/2024 12:09:47",
      "content": "<p>Hi, I am not sure if I understand it correctly, but isn't that the fact that MONTH and WEEK_NUM columns can be directly calculated from date_decision?<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2Ff9916bef40ff638c6d0322a90c569e4b%2F2024-04-04%20144508.png?generation=1712213197463571&amp;alt=media\" alt=\"MONTH\"><br>\nif so, what is the meaning of setting WEEK_NUM and MONTH as constant in the given test data?<br>\nWhat's more, do the date_decision corresponds to real date when the decision is made? I am thinking about important relevant US interest rate and other data to facilitate my modeling?<br>\nCan anyone help me out? many thanks.</p>",
      "rawMarkdown": "Hi, I am not sure if I understand it correctly, but isn't that the fact that MONTH and WEEK_NUM columns can be directly calculated from date_decision?\n![MONTH](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2Ff9916bef40ff638c6d0322a90c569e4b%2F2024-04-04%20144508.png?generation=1712213197463571&alt=media)\nif so, what is the meaning of setting WEEK_NUM and MONTH as constant in the given test data?\nWhat's more, do the date_decision corresponds to real date when the decision is made? I am thinking about important relevant US interest rate and other data to facilitate my modeling?\nCan anyone help me out? many thanks.",
      "votes": null
    },
    {
      "id": "2735511",
      "postDate": "04/04/2024 18:37:14",
      "content": "<p>Here the all information that u need <a href=\"https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/482474\" target=\"_blank\">https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/482474</a></p>",
      "rawMarkdown": "Here the all information that u need https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/482474",
      "votes": null
    },
    {
      "id": "2736352",
      "postDate": "04/05/2024 06:38:13",
      "content": "<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F4cace7ebbebf6b784fbea64d80a61c64%2F2024-04-05%20143539.png?generation=1712299010635476&amp;alt=media\"><br>\nHi, i read through the post you mentioned but i am still confused. Like what's showed above, the MONTH column can be 100% calculated from the date_decision column in the train data. If that's the case, why can't i calculate the MONTH column myself in the test data?</p>",
      "rawMarkdown": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F4cace7ebbebf6b784fbea64d80a61c64%2F2024-04-05%20143539.png?generation=1712299010635476&alt=media)\nHi, i read through the post you mentioned but i am still confused. Like what's showed above, the MONTH column can be 100% calculated from the date_decision column in the train data. If that's the case, why can't i calculate the MONTH column myself in the test data?",
      "votes": null
    },
    {
      "id": "2736366",
      "postDate": "04/05/2024 06:48:33",
      "content": "<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F0aa618f81d63c5b3bb6f78563093647a%2F2024-04-05%20144653.png?generation=1712299688132076&amp;alt=media\"><br>\nthe WEEK_NUM can also be 100% calculated from date_decision too.</p>",
      "rawMarkdown": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F0aa618f81d63c5b3bb6f78563093647a%2F2024-04-05%20144653.png?generation=1712299688132076&alt=media)\nthe WEEK_NUM can also be 100% calculated from date_decision too.",
      "votes": null
    },
    {
      "id": "2736562",
      "postDate": "04/05/2024 09:15:02",
      "content": "<p>Read it carefully.<br>\n<strong>Column date_decision is no longer the same. The remaining data were also transformed</strong></p>",
      "rawMarkdown": "Read it carefully.\n**Column date_decision is no longer the same. The remaining data were also transformed**",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2735511,
      "author_name": "ignatgotin",
      "author_url": "",
      "post_date": "04/04/2024 18:37:14",
      "content": "<p>Here the all information that u need <a href=\"https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/482474\" target=\"_blank\">https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/482474</a></p>",
      "votes": null,
      "replies": [
        {
          "id": 2736352,
          "author_name": "jasonhuang20040812",
          "author_url": "",
          "post_date": "04/05/2024 06:38:13",
          "content": "<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F4cace7ebbebf6b784fbea64d80a61c64%2F2024-04-05%20143539.png?generation=1712299010635476&amp;alt=media\"><br>\nHi, i read through the post you mentioned but i am still confused. Like what's showed above, the MONTH column can be 100% calculated from the date_decision column in the train data. If that's the case, why can't i calculate the MONTH column myself in the test data?</p>",
          "votes": null,
          "replies": [
            {
              "id": 2736366,
              "author_name": "jasonhuang20040812",
              "author_url": "",
              "post_date": "04/05/2024 06:48:33",
              "content": "<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F0aa618f81d63c5b3bb6f78563093647a%2F2024-04-05%20144653.png?generation=1712299688132076&amp;alt=media\"><br>\nthe WEEK_NUM can also be 100% calculated from date_decision too.</p>",
              "votes": null,
              "replies": []
            },
            {
              "id": 2736562,
              "author_name": "ignatgotin",
              "author_url": "",
              "post_date": "04/05/2024 09:15:02",
              "content": "<p>Read it carefully.<br>\n<strong>Column date_decision is no longer the same. The remaining data were also transformed</strong></p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2734930": "Hi, I am not sure if I understand it correctly, but isn't that the fact that MONTH and WEEK_NUM columns can be directly calculated from date_decision?\n![MONTH](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2Ff9916bef40ff638c6d0322a90c569e4b%2F2024-04-04%20144508.png?generation=1712213197463571&alt=media)\nif so, what is the meaning of setting WEEK_NUM and MONTH as constant in the given test data?\nWhat's more, do the date_decision corresponds to real date when the decision is made? I am thinking about important relevant US interest rate and other data to facilitate my modeling?\nCan anyone help me out? many thanks.",
    "2735511": "Here the all information that u need https://www.kaggle.com/competitions/home-credit-credit-risk-model-stability/discussion/482474",
    "2736352": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F4cace7ebbebf6b784fbea64d80a61c64%2F2024-04-05%20143539.png?generation=1712299010635476&alt=media)\nHi, i read through the post you mentioned but i am still confused. Like what's showed above, the MONTH column can be 100% calculated from the date_decision column in the train data. If that's the case, why can't i calculate the MONTH column myself in the test data?",
    "2736366": "![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F18899077%2F0aa618f81d63c5b3bb6f78563093647a%2F2024-04-05%20144653.png?generation=1712299688132076&alt=media)\nthe WEEK_NUM can also be 100% calculated from date_decision too.",
    "2736562": "Read it carefully.\n**Column date_decision is no longer the same. The remaining data were also transformed**"
  },
  "source": "meta"
}