{
  "id": 547262,
  "title": "Feature in test data are \"0\"",
  "url": "/competitions/jane-street-real-time-market-data-forecasting/discussion/547262",
  "author_name": "",
  "post_date": "2024-11-20T15:24:59.699383500Z",
  "votes": 1,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I just follow the <a href=\"https://www.kaggle.com/code/ryanholbrook/jane-street-rmf-demo-submission\" target=\"_blank\">demo</a> to import the test data, but I noticed that all the features in the test dataset are zero. Is this expected behavior, or could I be missing something?</p>\n<pre><code>inference_server = kaggle_evaluation.jane_street_inference_server.JSInferenceServer(predict)\n\n os.getenv():\n    inference_server.serve()\n:\n    inference_server.run_local_gateway(\n        (\n            ,\n            ,\n        )\n    )\n</code></pre>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F3c022a4779b0743c76e1be849ba42bc1%2F.jpg?generation=1732116269928384&amp;alt=media\" alt=\"\"></p>",
  "messages": [
    {
      "id": "3050828",
      "postDate": "11/20/2024 15:24:59",
      "content": "<p>I just follow the <a href=\"https://www.kaggle.com/code/ryanholbrook/jane-street-rmf-demo-submission\" target=\"_blank\">demo</a> to import the test data, but I noticed that all the features in the test dataset are zero. Is this expected behavior, or could I be missing something?</p>\n<pre><code>inference_server = kaggle_evaluation.jane_street_inference_server.JSInferenceServer(predict)\n\n os.getenv():\n    inference_server.serve()\n:\n    inference_server.run_local_gateway(\n        (\n            ,\n            ,\n        )\n    )\n</code></pre>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F3c022a4779b0743c76e1be849ba42bc1%2F.jpg?generation=1732116269928384&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "I just follow the [demo](https://www.kaggle.com/code/ryanholbrook/jane-street-rmf-demo-submission) to import the test data, but I noticed that all the features in the test dataset are zero. Is this expected behavior, or could I be missing something?\n\n```python\ninference_server = kaggle_evaluation.jane_street_inference_server.JSInferenceServer(predict)\n\nif os.getenv('KAGGLE_IS_COMPETITION_RERUN'):\n    inference_server.serve()\nelse:\n    inference_server.run_local_gateway(\n        (\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/test.parquet',\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/lags.parquet',\n        )\n    )\n```\n\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F3c022a4779b0743c76e1be849ba42bc1%2F.jpg?generation=1732116269928384&alt=media)",
      "votes": null
    },
    {
      "id": "3051084",
      "postDate": "11/20/2024 20:12:41",
      "content": "<p>This is a dummy test set batch to test your code <a href=\"https://www.kaggle.com/abcd334\" target=\"_blank\">@abcd334</a> <br>\nThe actual test set is hidden and will be served when you will submit to the leaderboard.</p>",
      "rawMarkdown": "This is a dummy test set batch to test your code @abcd334 \nThe actual test set is hidden and will be served when you will submit to the leaderboard.",
      "votes": null
    },
    {
      "id": "3051798",
      "postDate": "11/21/2024 16:29:05",
      "content": "<p>If I combine test and lags, will the features in test remain as zeros since I am not allowed to modify the original test data?</p>",
      "rawMarkdown": "If I combine test and lags, will the features in test remain as zeros since I am not allowed to modify the original test data?",
      "votes": null
    },
    {
      "id": "3052303",
      "postDate": "11/22/2024 09:06:57",
      "content": "<p>No, and they won't be zeros anyway, that's the point.</p>\n<p><code>test.parquet</code> and <code>lags.parquet</code> are provided simply to help us structure our data. What you see in these files in schema is what will be served up to the <code>predict</code> function during the test phase.</p>\n<p>The <code>predict</code> function will be run by the server on <code>test</code> and <code>lags</code> polars dataframes supplied to the function by the server, in the same schema as the parquet files, except populated with the actual data. So right now even in the public test set, they will not be zeros.</p>",
      "rawMarkdown": "No, and they won't be zeros anyway, that's the point.\n\n`test.parquet` and `lags.parquet` are provided simply to help us structure our data. What you see in these files in schema is what will be served up to the `predict` function during the test phase.\n\nThe `predict` function will be run by the server on `test` and `lags` polars dataframes supplied to the function by the server, in the same schema as the parquet files, except populated with the actual data. So right now even in the public test set, they will not be zeros.",
      "votes": null
    },
    {
      "id": "3054478",
      "postDate": "11/24/2024 17:47:54",
      "content": "<p>I submitted my solution, but it keeps returning a \"Submission Format Error\". Could you please help me understand what might be causing this issue?</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F03cc6afc33767e109dc1d14a255b7575%2Fsubmit.jpg?generation=1732470258231028&amp;alt=media\" alt=\"\"></p>",
      "rawMarkdown": "I submitted my solution, but it keeps returning a \"Submission Format Error\". Could you please help me understand what might be causing this issue?\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F03cc6afc33767e109dc1d14a255b7575%2Fsubmit.jpg?generation=1732470258231028&alt=media)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3051084,
      "author_name": "ravi20076",
      "author_url": "",
      "post_date": "11/20/2024 20:12:41",
      "content": "<p>This is a dummy test set batch to test your code <a href=\"https://www.kaggle.com/abcd334\" target=\"_blank\">@abcd334</a> <br>\nThe actual test set is hidden and will be served when you will submit to the leaderboard.</p>",
      "votes": null,
      "replies": [
        {
          "id": 3051798,
          "author_name": "abcd334",
          "author_url": "",
          "post_date": "11/21/2024 16:29:05",
          "content": "<p>If I combine test and lags, will the features in test remain as zeros since I am not allowed to modify the original test data?</p>",
          "votes": null,
          "replies": [
            {
              "id": 3052303,
              "author_name": "abstrphil",
              "author_url": "",
              "post_date": "11/22/2024 09:06:57",
              "content": "<p>No, and they won't be zeros anyway, that's the point.</p>\n<p><code>test.parquet</code> and <code>lags.parquet</code> are provided simply to help us structure our data. What you see in these files in schema is what will be served up to the <code>predict</code> function during the test phase.</p>\n<p>The <code>predict</code> function will be run by the server on <code>test</code> and <code>lags</code> polars dataframes supplied to the function by the server, in the same schema as the parquet files, except populated with the actual data. So right now even in the public test set, they will not be zeros.</p>",
              "votes": null,
              "replies": []
            }
          ]
        },
        {
          "id": 3054478,
          "author_name": "abcd334",
          "author_url": "",
          "post_date": "11/24/2024 17:47:54",
          "content": "<p>I submitted my solution, but it keeps returning a \"Submission Format Error\". Could you please help me understand what might be causing this issue?</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F03cc6afc33767e109dc1d14a255b7575%2Fsubmit.jpg?generation=1732470258231028&amp;alt=media\" alt=\"\"></p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3050828": "I just follow the [demo](https://www.kaggle.com/code/ryanholbrook/jane-street-rmf-demo-submission) to import the test data, but I noticed that all the features in the test dataset are zero. Is this expected behavior, or could I be missing something?\n\n```python\ninference_server = kaggle_evaluation.jane_street_inference_server.JSInferenceServer(predict)\n\nif os.getenv('KAGGLE_IS_COMPETITION_RERUN'):\n    inference_server.serve()\nelse:\n    inference_server.run_local_gateway(\n        (\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/test.parquet',\n            '/kaggle/input/jane-street-real-time-market-data-forecasting/lags.parquet',\n        )\n    )\n```\n\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F3c022a4779b0743c76e1be849ba42bc1%2F.jpg?generation=1732116269928384&alt=media)",
    "3051084": "This is a dummy test set batch to test your code @abcd334 \nThe actual test set is hidden and will be served when you will submit to the leaderboard.",
    "3051798": "If I combine test and lags, will the features in test remain as zeros since I am not allowed to modify the original test data?",
    "3052303": "No, and they won't be zeros anyway, that's the point.\n\n`test.parquet` and `lags.parquet` are provided simply to help us structure our data. What you see in these files in schema is what will be served up to the `predict` function during the test phase.\n\nThe `predict` function will be run by the server on `test` and `lags` polars dataframes supplied to the function by the server, in the same schema as the parquet files, except populated with the actual data. So right now even in the public test set, they will not be zeros.",
    "3054478": "I submitted my solution, but it keeps returning a \"Submission Format Error\". Could you please help me understand what might be causing this issue?\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F22676567%2F03cc6afc33767e109dc1d14a255b7575%2Fsubmit.jpg?generation=1732470258231028&alt=media)"
  },
  "source": "meta"
}