{
  "id": 268131,
  "title": "submission problem",
  "url": "/competitions/rsna-miccai-brain-tumor-radiogenomic-classification/discussion/268131",
  "author_name": "",
  "post_date": "2021-08-26T06:44:25.938905200Z",
  "votes": 3,
  "comment_count": 5,
  "views": 0,
  "content": "<p>Hello,</p>\n<p>For some reason, I can not make a submission without \"Submission Scoring Error\". <br>\nJust to try to find the problem I made a simple notebook:  <a href=\"https://www.kaggle.com/sajozsattila/test-submission\" target=\"_blank\">https://www.kaggle.com/sajozsattila/test-submission</a></p>\n<p>This loads the '../input/rsna-miccai-brain-tumor-radiogenomic-classification/sample_submission.csv'  to df_example and my testfile in pandas df. They are equal (df.equals(df_example) == True), however, if the df_example is a processing but the other one is not.  Any idea what going on? </p>",
  "messages": [
    {
      "id": "1491098",
      "postDate": "08/26/2021 06:44:25",
      "content": "<p>Hello,</p>\n<p>For some reason, I can not make a submission without \"Submission Scoring Error\". <br>\nJust to try to find the problem I made a simple notebook:  <a href=\"https://www.kaggle.com/sajozsattila/test-submission\" target=\"_blank\">https://www.kaggle.com/sajozsattila/test-submission</a></p>\n<p>This loads the '../input/rsna-miccai-brain-tumor-radiogenomic-classification/sample_submission.csv'  to df_example and my testfile in pandas df. They are equal (df.equals(df_example) == True), however, if the df_example is a processing but the other one is not.  Any idea what going on? </p>",
      "rawMarkdown": "Hello,\n\nFor some reason, I can not make a submission without \"Submission Scoring Error\". \nJust to try to find the problem I made a simple notebook:  https://www.kaggle.com/sajozsattila/test-submission\n\nThis loads the '../input/rsna-miccai-brain-tumor-radiogenomic-classification/sample_submission.csv'  to df_example and my testfile in pandas df. They are equal (df.equals(df_example) == True), however, if the df_example is a processing but the other one is not.  Any idea what going on?",
      "votes": null
    },
    {
      "id": "1491136",
      "postDate": "08/26/2021 07:19:10",
      "content": "<p>I think it's because when we submit the notebook for scoring, the <code>sample_submission.csv</code> file is replaced with another version that consists of all patients in the test set (instead of just 87 patients in the public test set).</p>",
      "rawMarkdown": "I think it's because when we submit the notebook for scoring, the `sample_submission.csv` file is replaced with another version that consists of all patients in the test set (instead of just 87 patients in the public test set).",
      "votes": null
    },
    {
      "id": "1491827",
      "postDate": "08/26/2021 16:42:15",
      "content": "<p>Probably I misunderstanding something, but I thought I need to produce an output prediction for the 87 patients (the submission.csv).<br>\n If not what is need to be there actually?  A model? Called what? </p>",
      "rawMarkdown": "Probably I misunderstanding something, but I thought I need to produce an output prediction for the 87 patients (the submission.csv).\n If not what is need to be there actually?  A model? Called what?",
      "votes": null
    },
    {
      "id": "1491859",
      "postDate": "08/26/2021 17:05:07",
      "content": "<p>The 87 patients set is only the public part of the test set. You need to produce an output prediction for the whole test set, including the hidden/private part.</p>",
      "rawMarkdown": "The 87 patients set is only the public part of the test set. You need to produce an output prediction for the whole test set, including the hidden/private part.",
      "votes": null
    },
    {
      "id": "1493426",
      "postDate": "08/27/2021 21:41:58",
      "content": "<p>This submission still does not work for me. Now I get \"No output file found\".  which is interesting as I can see:<br>\n<code>print(os.listdir(\"./\"))</code> is <code>['data_1.4_predict', '__notebook__.ipynb', 'data_1.4_filtered', 'datalabel_prediction_1.4.4_t2.csv', 'data_1.4', 'model_1.4.4_t2.pth', 'submission2.csv', 'datalabel_hastomur_1.4.4_t2.csv', 'submission.csv']</code> </p>\n<p>And if I open the submission.csv I think it is in the right format:</p>\n<pre><code>BraTS21ID,MGMT_value\n1,0.26785714285714285\n13,0.5575539568345323\n15,0.5855513307984791\n27,0.6379310344827587\n...\n</code></pre>",
      "rawMarkdown": "This submission still does not work for me. Now I get \"No output file found\".  which is interesting as I can see:\n`print(os.listdir(\"./\"))` is `['data_1.4_predict', '__notebook__.ipynb', 'data_1.4_filtered', 'datalabel_prediction_1.4.4_t2.csv', 'data_1.4', 'model_1.4.4_t2.pth', 'submission2.csv', 'datalabel_hastomur_1.4.4_t2.csv', 'submission.csv']` \n\nAnd if I open the submission.csv I think it is in the right format:\n```\nBraTS21ID,MGMT_value\n1,0.26785714285714285\n13,0.5575539568345323\n15,0.5855513307984791\n27,0.6379310344827587\n...\n```",
      "votes": null
    },
    {
      "id": "1501605",
      "postDate": "09/03/2021 11:00:22",
      "content": "<p>So the problem looks like connected to some working directory on my notebook. When I have deleted all of these tmp directories, it started to see the submission.csv.</p>",
      "rawMarkdown": "So the problem looks like connected to some working directory on my notebook. When I have deleted all of these tmp directories, it started to see the submission.csv.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1491136,
      "author_name": "nanguyen",
      "author_url": "",
      "post_date": "08/26/2021 07:19:10",
      "content": "<p>I think it's because when we submit the notebook for scoring, the <code>sample_submission.csv</code> file is replaced with another version that consists of all patients in the test set (instead of just 87 patients in the public test set).</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1491827,
      "author_name": "sajozsattila",
      "author_url": "",
      "post_date": "08/26/2021 16:42:15",
      "content": "<p>Probably I misunderstanding something, but I thought I need to produce an output prediction for the 87 patients (the submission.csv).<br>\n If not what is need to be there actually?  A model? Called what? </p>",
      "votes": null,
      "replies": [
        {
          "id": 1491859,
          "author_name": "nanguyen",
          "author_url": "",
          "post_date": "08/26/2021 17:05:07",
          "content": "<p>The 87 patients set is only the public part of the test set. You need to produce an output prediction for the whole test set, including the hidden/private part.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1493426,
      "author_name": "sajozsattila",
      "author_url": "",
      "post_date": "08/27/2021 21:41:58",
      "content": "<p>This submission still does not work for me. Now I get \"No output file found\".  which is interesting as I can see:<br>\n<code>print(os.listdir(\"./\"))</code> is <code>['data_1.4_predict', '__notebook__.ipynb', 'data_1.4_filtered', 'datalabel_prediction_1.4.4_t2.csv', 'data_1.4', 'model_1.4.4_t2.pth', 'submission2.csv', 'datalabel_hastomur_1.4.4_t2.csv', 'submission.csv']</code> </p>\n<p>And if I open the submission.csv I think it is in the right format:</p>\n<pre><code>BraTS21ID,MGMT_value\n1,0.26785714285714285\n13,0.5575539568345323\n15,0.5855513307984791\n27,0.6379310344827587\n...\n</code></pre>",
      "votes": null,
      "replies": [
        {
          "id": 1501605,
          "author_name": "sajozsattila",
          "author_url": "",
          "post_date": "09/03/2021 11:00:22",
          "content": "<p>So the problem looks like connected to some working directory on my notebook. When I have deleted all of these tmp directories, it started to see the submission.csv.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1491098": "Hello,\n\nFor some reason, I can not make a submission without \"Submission Scoring Error\". \nJust to try to find the problem I made a simple notebook:  https://www.kaggle.com/sajozsattila/test-submission\n\nThis loads the '../input/rsna-miccai-brain-tumor-radiogenomic-classification/sample_submission.csv'  to df_example and my testfile in pandas df. They are equal (df.equals(df_example) == True), however, if the df_example is a processing but the other one is not.  Any idea what going on?",
    "1491136": "I think it's because when we submit the notebook for scoring, the `sample_submission.csv` file is replaced with another version that consists of all patients in the test set (instead of just 87 patients in the public test set).",
    "1491827": "Probably I misunderstanding something, but I thought I need to produce an output prediction for the 87 patients (the submission.csv).\n If not what is need to be there actually?  A model? Called what?",
    "1491859": "The 87 patients set is only the public part of the test set. You need to produce an output prediction for the whole test set, including the hidden/private part.",
    "1493426": "This submission still does not work for me. Now I get \"No output file found\".  which is interesting as I can see:\n`print(os.listdir(\"./\"))` is `['data_1.4_predict', '__notebook__.ipynb', 'data_1.4_filtered', 'datalabel_prediction_1.4.4_t2.csv', 'data_1.4', 'model_1.4.4_t2.pth', 'submission2.csv', 'datalabel_hastomur_1.4.4_t2.csv', 'submission.csv']` \n\nAnd if I open the submission.csv I think it is in the right format:\n```\nBraTS21ID,MGMT_value\n1,0.26785714285714285\n13,0.5575539568345323\n15,0.5855513307984791\n27,0.6379310344827587\n...\n```",
    "1501605": "So the problem looks like connected to some working directory on my notebook. When I have deleted all of these tmp directories, it started to see the submission.csv."
  },
  "source": "meta"
}