{
  "id": 548309,
  "title": "some questions about time_id",
  "url": "/competitions/jane-street-real-time-market-data-forecasting/discussion/548309",
  "author_name": "",
  "post_date": "2024-11-26T06:00:39.103220100Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hello everyone, I'm participating in the competition in the field of quants for the first time, so I'm a little confused about some of the parts. If I choose the attention layer as my model and choose T-100 to T-1 as the training set to predict the data at time T, then there won't necessarily be a period of 100 in the test set, right? And is the flag ID in the test set exactly the same as that in the training set?</p>",
  "messages": [
    {
      "id": "3055783",
      "postDate": "11/26/2024 06:00:39",
      "content": "<p>Hello everyone, I'm participating in the competition in the field of quants for the first time, so I'm a little confused about some of the parts. If I choose the attention layer as my model and choose T-100 to T-1 as the training set to predict the data at time T, then there won't necessarily be a period of 100 in the test set, right? And is the flag ID in the test set exactly the same as that in the training set?</p>",
      "rawMarkdown": "Hello everyone, I'm participating in the competition in the field of quants for the first time, so I'm a little confused about some of the parts. If I choose the attention layer as my model and choose T-100 to T-1 as the training set to predict the data at time T, then there won't necessarily be a period of 100 in the test set, right? And is the flag ID in the test set exactly the same as that in the training set?",
      "votes": null
    },
    {
      "id": "3056652",
      "postDate": "11/27/2024 08:37:38",
      "content": "<p>This task is more like tabular data. What can be received during prediction is only the data of the current time step and the lag of the previous date (perhaps you can store the previous data in the prediction function to use the time series correlation model, but I have not tried it. )</p>",
      "rawMarkdown": "This task is more like tabular data. What can be received during prediction is only the data of the current time step and the lag of the previous date (perhaps you can store the previous data in the prediction function to use the time series correlation model, but I have not tried it. )",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3056652,
      "author_name": "dearluna",
      "author_url": "",
      "post_date": "11/27/2024 08:37:38",
      "content": "<p>This task is more like tabular data. What can be received during prediction is only the data of the current time step and the lag of the previous date (perhaps you can store the previous data in the prediction function to use the time series correlation model, but I have not tried it. )</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3055783": "Hello everyone, I'm participating in the competition in the field of quants for the first time, so I'm a little confused about some of the parts. If I choose the attention layer as my model and choose T-100 to T-1 as the training set to predict the data at time T, then there won't necessarily be a period of 100 in the test set, right? And is the flag ID in the test set exactly the same as that in the training set?",
    "3056652": "This task is more like tabular data. What can be received during prediction is only the data of the current time step and the lag of the previous date (perhaps you can store the previous data in the prediction function to use the time series correlation model, but I have not tried it. )"
  },
  "source": "meta"
}