{
  "id": 469065,
  "title": "Submission Scoring Error",
  "url": "/competitions/hms-harmful-brain-activity-classification/discussion/469065",
  "author_name": "",
  "post_date": "2024-01-19T00:17:16.965125200Z",
  "votes": 7,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hello guys,<br>\nI have been working on this bug for 2 days now. I did 9 submissions and all returned the same error \"Submission Scoring Error\". </p>\n<p>The following is the content of the my submission file</p>\n<table>\n<thead>\n<tr>\n<th></th>\n<th>eeg_id</th>\n<th>seizure_vote</th>\n<th>lpd_vote</th>\n<th>gpd_vote</th>\n<th>lrda_vote</th>\n<th>grda_vote</th>\n<th>other_vote</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>0</td>\n<td>3911565283</td>\n<td>0.169783</td>\n<td>0.163736</td>\n<td>0.159434</td>\n<td>0.161185</td>\n<td>0.164098</td>\n<td>0.181764</td>\n</tr>\n</tbody>\n</table>\n<p>The datatype of eeg_id is numeric, and all of the *_vote values add up to one. What would be the problem?</p>\n<p>Thanks in advance</p>",
  "messages": [
    {
      "id": "2608603",
      "postDate": "01/19/2024 00:17:16",
      "content": "<p>Hello guys,<br>\nI have been working on this bug for 2 days now. I did 9 submissions and all returned the same error \"Submission Scoring Error\". </p>\n<p>The following is the content of the my submission file</p>\n<table>\n<thead>\n<tr>\n<th></th>\n<th>eeg_id</th>\n<th>seizure_vote</th>\n<th>lpd_vote</th>\n<th>gpd_vote</th>\n<th>lrda_vote</th>\n<th>grda_vote</th>\n<th>other_vote</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>0</td>\n<td>3911565283</td>\n<td>0.169783</td>\n<td>0.163736</td>\n<td>0.159434</td>\n<td>0.161185</td>\n<td>0.164098</td>\n<td>0.181764</td>\n</tr>\n</tbody>\n</table>\n<p>The datatype of eeg_id is numeric, and all of the *_vote values add up to one. What would be the problem?</p>\n<p>Thanks in advance</p>",
      "rawMarkdown": "Hello guys,\nI have been working on this bug for 2 days now. I did 9 submissions and all returned the same error \"Submission Scoring Error\". \n\nThe following is the content of the my submission file\n| |  eeg_id  | seizure_vote  | lpd_vote  | gpd_vote  | lrda_vote  | grda_vote  | other_vote |  \n| --- | --- | --- | --- | --- | --- | --- | --- | \n| 0 | 3911565283 | 0.169783 | 0.163736 | 0.159434 | 0.161185 | 0.164098 | 0.181764| \n\nThe datatype of eeg_id is numeric, and all of the *_vote values add up to one. What would be the problem?\n\nThanks in advance",
      "votes": null
    },
    {
      "id": "2611741",
      "postDate": "01/20/2024 23:57:03",
      "content": "<p>Hi, I am having the same problem right now, what I've tried - converting all the predictions in float64 and renormalize them (divide by sum after expanding the precision). Didn't work for me, but maybe will work for you. Also it seems that you have the row number as an index in your dataframe, do you save your submission with it or not? Like <code>df.to_csv(index=False)</code>, because the example submission has no index in it (or the eeg_id is the index, yet no rows enumeration).</p>",
      "rawMarkdown": "Hi, I am having the same problem right now, what I've tried - converting all the predictions in float64 and renormalize them (divide by sum after expanding the precision). Didn't work for me, but maybe will work for you. Also it seems that you have the row number as an index in your dataframe, do you save your submission with it or not? Like `df.to_csv(index=False)`, because the example submission has no index in it (or the eeg_id is the index, yet no rows enumeration).",
      "votes": null
    },
    {
      "id": "2611767",
      "postDate": "01/21/2024 00:56:55",
      "content": "<p>I figured it out. Actually I was designing my test code to account for only 1 row, since the test.csv contains only one sample, however, it seems that when you submit the code, it runs on a different <code>test.csv</code> . Thus, you need to make your testing code more generic such that it can handle more than one test case.</p>\n<p>I believe that this error is due to <code>nan</code> values in the submission dataframe that I was creating. Kindly let me know if this information is helpful for you as well.</p>",
      "rawMarkdown": "I figured it out. Actually I was designing my test code to account for only 1 row, since the test.csv contains only one sample, however, it seems that when you submit the code, it runs on a different `test.csv` . Thus, you need to make your testing code more generic such that it can handle more than one test case.\n\nI believe that this error is due to `nan` values in the submission dataframe that I was creating. Kindly let me know if this information is helpful for you as well.",
      "votes": null
    },
    {
      "id": "2635668",
      "postDate": "02/04/2024 14:49:21",
      "content": "<p>I had the same issue. turns out the competition is using some hidden test data for scoring</p>",
      "rawMarkdown": "I had the same issue. turns out the competition is using some hidden test data for scoring",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2611741,
      "author_name": "kst179",
      "author_url": "",
      "post_date": "01/20/2024 23:57:03",
      "content": "<p>Hi, I am having the same problem right now, what I've tried - converting all the predictions in float64 and renormalize them (divide by sum after expanding the precision). Didn't work for me, but maybe will work for you. Also it seems that you have the row number as an index in your dataframe, do you save your submission with it or not? Like <code>df.to_csv(index=False)</code>, because the example submission has no index in it (or the eeg_id is the index, yet no rows enumeration).</p>",
      "votes": null,
      "replies": [
        {
          "id": 2611767,
          "author_name": "silk1100",
          "author_url": "",
          "post_date": "01/21/2024 00:56:55",
          "content": "<p>I figured it out. Actually I was designing my test code to account for only 1 row, since the test.csv contains only one sample, however, it seems that when you submit the code, it runs on a different <code>test.csv</code> . Thus, you need to make your testing code more generic such that it can handle more than one test case.</p>\n<p>I believe that this error is due to <code>nan</code> values in the submission dataframe that I was creating. Kindly let me know if this information is helpful for you as well.</p>",
          "votes": null,
          "replies": [
            {
              "id": 2635668,
              "author_name": "eouedraogo4",
              "author_url": "",
              "post_date": "02/04/2024 14:49:21",
              "content": "<p>I had the same issue. turns out the competition is using some hidden test data for scoring</p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2608603": "Hello guys,\nI have been working on this bug for 2 days now. I did 9 submissions and all returned the same error \"Submission Scoring Error\". \n\nThe following is the content of the my submission file\n| |  eeg_id  | seizure_vote  | lpd_vote  | gpd_vote  | lrda_vote  | grda_vote  | other_vote |  \n| --- | --- | --- | --- | --- | --- | --- | --- | \n| 0 | 3911565283 | 0.169783 | 0.163736 | 0.159434 | 0.161185 | 0.164098 | 0.181764| \n\nThe datatype of eeg_id is numeric, and all of the *_vote values add up to one. What would be the problem?\n\nThanks in advance",
    "2611741": "Hi, I am having the same problem right now, what I've tried - converting all the predictions in float64 and renormalize them (divide by sum after expanding the precision). Didn't work for me, but maybe will work for you. Also it seems that you have the row number as an index in your dataframe, do you save your submission with it or not? Like `df.to_csv(index=False)`, because the example submission has no index in it (or the eeg_id is the index, yet no rows enumeration).",
    "2611767": "I figured it out. Actually I was designing my test code to account for only 1 row, since the test.csv contains only one sample, however, it seems that when you submit the code, it runs on a different `test.csv` . Thus, you need to make your testing code more generic such that it can handle more than one test case.\n\nI believe that this error is due to `nan` values in the submission dataframe that I was creating. Kindly let me know if this information is helpful for you as well.",
    "2635668": "I had the same issue. turns out the competition is using some hidden test data for scoring"
  },
  "source": "meta"
}