{
  "id": 466705,
  "title": "train.csv is empty",
  "url": "/competitions/hms-harmful-brain-activity-classification/discussion/466705",
  "author_name": "",
  "post_date": "2024-01-09T17:15:26.385545700Z",
  "votes": 63,
  "comment_count": 11,
  "views": 0,
  "content": "<p>This looks like an interesting competition and timing is great. I just started downloading the data but I noticed that train.csv is empty. </p>",
  "messages": [
    {
      "id": "2594136",
      "postDate": "01/09/2024 17:15:26",
      "content": "<p>This looks like an interesting competition and timing is great. I just started downloading the data but I noticed that train.csv is empty. </p>",
      "rawMarkdown": "This looks like an interesting competition and timing is great. I just started downloading the data but I noticed that train.csv is empty.",
      "votes": null
    },
    {
      "id": "2594167",
      "postDate": "01/09/2024 17:35:17",
      "content": "<p>Indeed - same problem here.  And it isn't caused by the download - the website shows train.csv as a mere 209 bytes, which is clearly just the column headings as also reflected in the CSV viewer.</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2217262%2Fcb3797c247787a0656910082341a8c18%2Femptycsv.png?generation=1704821715707640&amp;alt=media\" alt=\"\"></p>\n<p>Same problem for test.csv too and the sample submission only has 1 row.  Mentioning <a href=\"https://www.kaggle.com/ashleychow\" target=\"_blank\">@ashleychow</a> as competition organizer to bring it to Kaggle staff attention.</p>",
      "rawMarkdown": "Indeed - same problem here.  And it isn't caused by the download - the website shows train.csv as a mere 209 bytes, which is clearly just the column headings as also reflected in the CSV viewer.\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2217262%2Fcb3797c247787a0656910082341a8c18%2Femptycsv.png?generation=1704821715707640&alt=media)\n\nSame problem for test.csv too and the sample submission only has 1 row.  Mentioning @ashleychow as competition organizer to bring it to Kaggle staff attention.",
      "votes": null
    },
    {
      "id": "2594168",
      "postDate": "01/09/2024 17:38:17",
      "content": "<p>Test and sample submission are probably correct since this is a code competition and they are placeholders.</p>",
      "rawMarkdown": "Test and sample submission are probably correct since this is a code competition and they are placeholders.",
      "votes": null
    },
    {
      "id": "2594182",
      "postDate": "01/09/2024 17:46:43",
      "content": "<p>Thank you for posting this we are looking into this now. </p>",
      "rawMarkdown": "Thank you for posting this we are looking into this now.",
      "votes": null
    },
    {
      "id": "2594220",
      "postDate": "01/09/2024 18:11:41",
      "content": "<p>Still empty, you can also check it in the notebook settings. The parquets are filled though</p>",
      "rawMarkdown": "Still empty, you can also check it in the notebook settings. The parquets are filled though",
      "votes": null
    },
    {
      "id": "2595279",
      "postDate": "01/10/2024 10:34:08",
      "content": "<p>The train data seems to be in the train_eegs folder in .parquet format (easy readable with pandas with <code>pd.read_parquet()</code> function). However I can't find the labels, which should be I guess in the (now empty) train.csv file.</p>",
      "rawMarkdown": "The train data seems to be in the train_eegs folder in .parquet format (easy readable with pandas with `pd.read_parquet()` function). However I can't find the labels, which should be I guess in the (now empty) train.csv file.",
      "votes": null
    },
    {
      "id": "2595500",
      "postDate": "01/10/2024 13:24:58",
      "content": "<p>is the competition going to be extended accordingly, as so far there isn't much one can do without the class annotations?…</p>",
      "rawMarkdown": "is the competition going to be extended accordingly, as so far there isn't much one can do without the class annotations?...",
      "votes": null
    },
    {
      "id": "2595897",
      "postDate": "01/10/2024 18:03:33",
      "content": "<p>Yes, the competition will be extended due to the technical difficulties experienced earlier.</p>",
      "rawMarkdown": "Yes, the competition will be extended due to the technical difficulties experienced earlier.",
      "votes": null
    },
    {
      "id": "2595981",
      "postDate": "01/10/2024 18:59:46",
      "content": "<p>Thanks for reporting this. I've just posted a patch.</p>",
      "rawMarkdown": "Thanks for reporting this. I've just posted a patch.",
      "votes": null
    },
    {
      "id": "2595995",
      "postDate": "01/10/2024 19:16:19",
      "content": "<p>Thank you for quick update !</p>",
      "rawMarkdown": "Thank you for quick update !",
      "votes": null
    },
    {
      "id": "2596654",
      "postDate": "01/11/2024 08:38:46",
      "content": "<p>Hello,<br>\nThanks a lot for the patch. It seems there are still some eegs in train_eegs folder that are not listed in the train.csv. When listing parquet files, I get 17300 while train csv file lists 17089 unique values for eeg_id. I think this notebook has already discovered that: <a href=\"https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet\" target=\"_blank\">https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet</a></p>",
      "rawMarkdown": "Hello,\nThanks a lot for the patch. It seems there are still some eegs in train_eegs folder that are not listed in the train.csv. When listing parquet files, I get 17300 while train csv file lists 17089 unique values for eeg_id. I think this notebook has already discovered that: [https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet](https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet)",
      "votes": null
    },
    {
      "id": "2598299",
      "postDate": "01/12/2024 09:19:50",
      "content": "<p>is this solved yet?</p>",
      "rawMarkdown": "is this solved yet?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2594167,
      "author_name": "andrewrrose",
      "author_url": "",
      "post_date": "01/09/2024 17:35:17",
      "content": "<p>Indeed - same problem here.  And it isn't caused by the download - the website shows train.csv as a mere 209 bytes, which is clearly just the column headings as also reflected in the CSV viewer.</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2217262%2Fcb3797c247787a0656910082341a8c18%2Femptycsv.png?generation=1704821715707640&amp;alt=media\" alt=\"\"></p>\n<p>Same problem for test.csv too and the sample submission only has 1 row.  Mentioning <a href=\"https://www.kaggle.com/ashleychow\" target=\"_blank\">@ashleychow</a> as competition organizer to bring it to Kaggle staff attention.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2594168,
          "author_name": "gunesevitan",
          "author_url": "",
          "post_date": "01/09/2024 17:38:17",
          "content": "<p>Test and sample submission are probably correct since this is a code competition and they are placeholders.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 2594220,
          "author_name": "maiernator",
          "author_url": "",
          "post_date": "01/09/2024 18:11:41",
          "content": "<p>Still empty, you can also check it in the notebook settings. The parquets are filled though</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 2594182,
      "author_name": "maggiemd",
      "author_url": "",
      "post_date": "01/09/2024 17:46:43",
      "content": "<p>Thank you for posting this we are looking into this now. </p>",
      "votes": null,
      "replies": [
        {
          "id": 2595500,
          "author_name": "iworeushankaonce",
          "author_url": "",
          "post_date": "01/10/2024 13:24:58",
          "content": "<p>is the competition going to be extended accordingly, as so far there isn't much one can do without the class annotations?…</p>",
          "votes": null,
          "replies": [
            {
              "id": 2595897,
              "author_name": "utkarshsagar1",
              "author_url": "",
              "post_date": "01/10/2024 18:03:33",
              "content": "<p>Yes, the competition will be extended due to the technical difficulties experienced earlier.</p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    },
    {
      "id": 2595279,
      "author_name": "jwackito",
      "author_url": "",
      "post_date": "01/10/2024 10:34:08",
      "content": "<p>The train data seems to be in the train_eegs folder in .parquet format (easy readable with pandas with <code>pd.read_parquet()</code> function). However I can't find the labels, which should be I guess in the (now empty) train.csv file.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2595981,
      "author_name": "sohier",
      "author_url": "",
      "post_date": "01/10/2024 18:59:46",
      "content": "<p>Thanks for reporting this. I've just posted a patch.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2595995,
          "author_name": "ttahara",
          "author_url": "",
          "post_date": "01/10/2024 19:16:19",
          "content": "<p>Thank you for quick update !</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 2596654,
      "author_name": "timotheguy",
      "author_url": "",
      "post_date": "01/11/2024 08:38:46",
      "content": "<p>Hello,<br>\nThanks a lot for the patch. It seems there are still some eegs in train_eegs folder that are not listed in the train.csv. When listing parquet files, I get 17300 while train csv file lists 17089 unique values for eeg_id. I think this notebook has already discovered that: <a href=\"https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet\" target=\"_blank\">https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet</a></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2598299,
      "author_name": "stacknishant",
      "author_url": "",
      "post_date": "01/12/2024 09:19:50",
      "content": "<p>is this solved yet?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2594136": "This looks like an interesting competition and timing is great. I just started downloading the data but I noticed that train.csv is empty.",
    "2594167": "Indeed - same problem here.  And it isn't caused by the download - the website shows train.csv as a mere 209 bytes, which is clearly just the column headings as also reflected in the CSV viewer.\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F2217262%2Fcb3797c247787a0656910082341a8c18%2Femptycsv.png?generation=1704821715707640&alt=media)\n\nSame problem for test.csv too and the sample submission only has 1 row.  Mentioning @ashleychow as competition organizer to bring it to Kaggle staff attention.",
    "2594168": "Test and sample submission are probably correct since this is a code competition and they are placeholders.",
    "2594182": "Thank you for posting this we are looking into this now.",
    "2594220": "Still empty, you can also check it in the notebook settings. The parquets are filled though",
    "2595279": "The train data seems to be in the train_eegs folder in .parquet format (easy readable with pandas with `pd.read_parquet()` function). However I can't find the labels, which should be I guess in the (now empty) train.csv file.",
    "2595500": "is the competition going to be extended accordingly, as so far there isn't much one can do without the class annotations?...",
    "2595897": "Yes, the competition will be extended due to the technical difficulties experienced earlier.",
    "2595981": "Thanks for reporting this. I've just posted a patch.",
    "2595995": "Thank you for quick update !",
    "2596654": "Hello,\nThanks a lot for the patch. It seems there are still some eegs in train_eegs folder that are not listed in the train.csv. When listing parquet files, I get 17300 while train csv file lists 17089 unique values for eeg_id. I think this notebook has already discovered that: [https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet](https://www.kaggle.com/code/seshurajup/missing-eeg-ids-in-train-csv-vs-train-eegs-parquet)",
    "2598299": "is this solved yet?"
  },
  "source": "meta"
}