{
  "id": 587957,
  "title": "Free external data to use",
  "url": "/competitions/drw-crypto-market-prediction/discussion/587957",
  "author_name": "yuanzhe zhou",
  "post_date": "2025-07-03T15:57:54.996000",
  "votes": 22,
  "comment_count": 14,
  "views": 0,
  "content": "<p>I'm not sure if it is already known to all, but free data is provided here: <br>\n<a href=\"https://github.com/binance/binance-public-data\" target=\"_blank\">https://github.com/binance/binance-public-data</a></p>",
  "messages": [
    {
      "id": 3240243,
      "postDate": "2025-07-03T15:57:54.997Z",
      "content": "<p>I'm not sure if it is already known to all, but free data is provided here: <br>\n<a href=\"https://github.com/binance/binance-public-data\" target=\"_blank\">https://github.com/binance/binance-public-data</a></p>",
      "rawMarkdown": "I'm not sure if it is already known to all, but free data is provided here: \n[https://github.com/binance/binance-public-data](https://github.com/binance/binance-public-data)\n\n",
      "votes": 22
    },
    {
      "id": 3240621,
      "postDate": "2025-07-04T04:20:23.470Z",
      "content": "<p>Thank you for your sharing! I have a question: since the timestamps in the test set have already been shuffled, how can we merge external data with the test set in this case?</p>",
      "rawMarkdown": "Thank you for your sharing! I have a question: since the timestamps in the test set have already been shuffled, how can we merge external data with the test set in this case?",
      "votes": 9
    },
    {
      "id": 3244742,
      "postDate": "2025-07-08T14:03:51.543Z",
      "content": "<p>I read one article on Kaggle. Author was thinking that we trying to predict BTC/USDT price change</p>",
      "rawMarkdown": "I read one article on Kaggle. Author was thinking that we trying to predict BTC/USDT price change",
      "votes": 1
    },
    {
      "id": 3244255,
      "postDate": "2025-07-08T02:49:03.907Z",
      "content": "<p>I believe one possible way to use the extra data is using the provided feature to predict your features, then predict them in the test dataset. <br>\nIn this way, no leakage is involved.</p>",
      "rawMarkdown": "I believe one possible way to use the extra data is using the provided feature to predict your features, then predict them in the test dataset. \nIn this way, no leakage is involved.",
      "votes": 1,
      "replies": [
        {
          "id": 3244268,
          "postDate": "2025-07-08T03:15:47.200Z",
          "content": "<p>Is the data we are use kline BTCUSD?</p>",
          "rawMarkdown": "Is the data we are use kline BTCUSD?"
        }
      ]
    },
    {
      "id": 3241498,
      "postDate": "2025-07-04T20:43:10.557Z",
      "content": "<p>Thanks for this amazing topic <a href=\"https://www.kaggle.com/yuanzhezhou\" target=\"_blank\">@yuanzhezhou</a></p>",
      "rawMarkdown": "Thanks for this amazing topic @yuanzhezhou",
      "votes": 2
    },
    {
      "id": 3243779,
      "postDate": "2025-07-07T14:12:44.997Z",
      "content": "<h2><strong>can we use it in drw competition?</strong></h2>",
      "rawMarkdown": "## **can we use it in drw competition?**"
    },
    {
      "id": 3242204,
      "postDate": "2025-07-05T17:18:47.100Z",
      "content": "<p>wonder how this data can be helpful.</p>",
      "rawMarkdown": "wonder how this data can be helpful."
    },
    {
      "id": 3244252,
      "postDate": "2025-07-08T02:45:34.850Z",
      "rawMarkdown": "",
      "isDeleted": true
    },
    {
      "id": 3241497,
      "postDate": "2025-07-04T20:41:51.493Z",
      "content": "<p>Thanks to share this <a href=\"https://www.kaggle.com/yuanzhezhou\" target=\"_blank\">@yuanzhezhou</a> </p>",
      "rawMarkdown": "Thanks to share this @yuanzhezhou ",
      "votes": 3
    },
    {
      "id": 3245544,
      "postDate": "2025-07-09T14:37:17.627Z",
      "content": "<p>Thanks for sharing…</p>",
      "rawMarkdown": "Thanks for sharing..."
    },
    {
      "id": 3241721,
      "postDate": "2025-07-05T05:40:40.510Z",
      "content": "<p>Thank you!</p>",
      "rawMarkdown": "Thank you!"
    },
    {
      "id": 3240898,
      "postDate": "2025-07-04T10:23:12.777Z",
      "content": "<p>Thank you!!</p>",
      "rawMarkdown": "Thank you!!"
    },
    {
      "id": 3240665,
      "postDate": "2025-07-04T05:29:02.380Z",
      "content": "<p>Thank you,mr. zhou,i am your fan.</p>",
      "rawMarkdown": "Thank you,mr. zhou,i am your fan."
    },
    {
      "id": 3240466,
      "postDate": "2025-07-03T20:34:37.683Z",
      "content": "<p>thank you!</p>",
      "rawMarkdown": "thank you!"
    }
  ],
  "comments": [
    {
      "id": 3240621,
      "author_name": "MLFINANCE",
      "author_url": "",
      "post_date": "2025-07-04T04:20:23.470000",
      "content": "<p>Thank you for your sharing! I have a question: since the timestamps in the test set have already been shuffled, how can we merge external data with the test set in this case?</p>",
      "votes": 9,
      "replies": []
    },
    {
      "id": 3244742,
      "author_name": "Dmitry Kiryukhin",
      "author_url": "",
      "post_date": "2025-07-08T14:03:51.543000",
      "content": "<p>I read one article on Kaggle. Author was thinking that we trying to predict BTC/USDT price change</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 3244255,
      "author_name": "yuanzhe zhou",
      "author_url": "",
      "post_date": "2025-07-08T02:49:03.907000",
      "content": "<p>I believe one possible way to use the extra data is using the provided feature to predict your features, then predict them in the test dataset. <br>\nIn this way, no leakage is involved.</p>",
      "votes": 1,
      "replies": [
        {
          "id": 3244268,
          "author_name": "paperxd",
          "author_url": "",
          "post_date": "2025-07-08T03:15:47.200000",
          "content": "<p>Is the data we are use kline BTCUSD?</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 3241498,
      "author_name": "HijabZahra",
      "author_url": "",
      "post_date": "2025-07-04T20:43:10.557000",
      "content": "<p>Thanks for this amazing topic <a href=\"https://www.kaggle.com/yuanzhezhou\" target=\"_blank\">@yuanzhezhou</a></p>",
      "votes": 2,
      "replies": []
    },
    {
      "id": 3243779,
      "author_name": "",
      "author_url": "",
      "post_date": "2025-07-07T14:12:44.997000",
      "content": "<h2><strong>can we use it in drw competition?</strong></h2>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 3242204,
      "author_name": "Artificial Gains",
      "author_url": "",
      "post_date": "2025-07-05T17:18:47.100000",
      "content": "<p>wonder how this data can be helpful.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 3244252,
      "author_name": "",
      "author_url": "",
      "post_date": "2025-07-08T02:45:34.850000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 3241497,
      "author_name": "Sadia Shahid latif",
      "author_url": "",
      "post_date": "2025-07-04T20:41:51.493000",
      "content": "<p>Thanks to share this <a href=\"https://www.kaggle.com/yuanzhezhou\" target=\"_blank\">@yuanzhezhou</a> </p>",
      "votes": 3,
      "replies": []
    },
    {
      "id": 3245544,
      "author_name": "Sarah Arshad",
      "author_url": "",
      "post_date": "2025-07-09T14:37:17.627000",
      "content": "<p>Thanks for sharing…</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 3241721,
      "author_name": "christopher obuchere",
      "author_url": "",
      "post_date": "2025-07-05T05:40:40.510000",
      "content": "<p>Thank you!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 3240898,
      "author_name": "Berke",
      "author_url": "",
      "post_date": "2025-07-04T10:23:12.777000",
      "content": "<p>Thank you!!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 3240665,
      "author_name": "sichao shen",
      "author_url": "",
      "post_date": "2025-07-04T05:29:02.380000",
      "content": "<p>Thank you,mr. zhou,i am your fan.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 3240466,
      "author_name": "debarupa :0",
      "author_url": "",
      "post_date": "2025-07-03T20:34:37.683000",
      "content": "<p>thank you!</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3240243": "I'm not sure if it is already known to all, but free data is provided here: \n[https://github.com/binance/binance-public-data](https://github.com/binance/binance-public-data)\n\n",
    "3240621": "Thank you for your sharing! I have a question: since the timestamps in the test set have already been shuffled, how can we merge external data with the test set in this case?",
    "3244742": "I read one article on Kaggle. Author was thinking that we trying to predict BTC/USDT price change",
    "3244255": "I believe one possible way to use the extra data is using the provided feature to predict your features, then predict them in the test dataset. \nIn this way, no leakage is involved.",
    "3241498": "Thanks for this amazing topic @yuanzhezhou",
    "3243779": "## **can we use it in drw competition?**",
    "3242204": "wonder how this data can be helpful.",
    "3244252": "",
    "3241497": "Thanks to share this @yuanzhezhou ",
    "3245544": "Thanks for sharing...",
    "3241721": "Thank you!",
    "3240898": "Thank you!!",
    "3240665": "Thank you,mr. zhou,i am your fan.",
    "3240466": "thank you!"
  }
}