{
  "id": 298340,
  "title": "[solved] Help! yet another \"Submission Scoring Error\"",
  "url": "/competitions/tensorflow-great-barrier-reef/discussion/298340",
  "author_name": "",
  "post_date": "2022-01-02T09:49:05.508986Z",
  "votes": 1,
  "comment_count": 8,
  "views": 0,
  "content": "<p>my notebook is at <a href=\"https://www.kaggle.com/hengck23/my-yolox-s-1440x2560\" target=\"_blank\">https://www.kaggle.com/hengck23/my-yolox-s-1440x2560</a><br>\nit has  \"Submission Scoring Error\"<br>\n( FYI:  this is an initial code to debug my pipeline. It uses yolox-s on input 1440x2560 trained with 5 epoches)</p>\n<p>in cell [6], I use run_check_detect_one() function to check that the model and prediction function are correct.</p>\n<p>cell [5] is the submission code that uses \" iter_test = env.iter_test() \" and results in the error after running for 40 min.<br>\ncan anyone help? thanks in advance!</p>",
  "messages": [
    {
      "id": "1635864",
      "postDate": "01/02/2022 09:49:05",
      "content": "<p>my notebook is at <a href=\"https://www.kaggle.com/hengck23/my-yolox-s-1440x2560\" target=\"_blank\">https://www.kaggle.com/hengck23/my-yolox-s-1440x2560</a><br>\nit has  \"Submission Scoring Error\"<br>\n( FYI:  this is an initial code to debug my pipeline. It uses yolox-s on input 1440x2560 trained with 5 epoches)</p>\n<p>in cell [6], I use run_check_detect_one() function to check that the model and prediction function are correct.</p>\n<p>cell [5] is the submission code that uses \" iter_test = env.iter_test() \" and results in the error after running for 40 min.<br>\ncan anyone help? thanks in advance!</p>",
      "rawMarkdown": "my notebook is at https://www.kaggle.com/hengck23/my-yolox-s-1440x2560\nit has  \"Submission Scoring Error\"\n( FYI:  this is an initial code to debug my pipeline. It uses yolox-s on input 1440x2560 trained with 5 epoches)\n\nin cell [6], I use run_check_detect_one() function to check that the model and prediction function are correct.\n\ncell [5] is the submission code that uses \" iter_test = env.iter_test() \" and results in the error after running for 40 min.\ncan anyone help? thanks in advance!",
      "votes": null
    },
    {
      "id": "1635870",
      "postDate": "01/02/2022 09:59:18",
      "content": "<p><a href=\"https://www.kaggle.com/hengck23\" target=\"_blank\">@hengck23</a> I ran into the same error a few days ago. I think this API works differently in <strong>public</strong> and <strong>private</strong>. I couldn't figure out what was wrong. Like you, my prediction and submission.csv format were all okay. After a bunch of experience here's what I found,<br>\nWe need to add <code>tensorflow-great-barrier-reef</code> to our system path even though we can import it w/o adding it in our system path.</p>\n<pre><code>import sys\nsys.path.append('../input/tensorflow-great-barrier-reef')\n</code></pre>",
      "rawMarkdown": "hengck23 I ran into the same error a few days ago. I think this API works differently in **public** and **private**. I couldn't figure out what was wrong. Like you, my prediction and submission.csv format were all okay. After a bunch of experience here's what I found,\nWe need to add `tensorflow-great-barrier-reef` to our system path even though we can import it w/o adding it in our system path.\n```\nimport sys\nsys.path.append('../input/tensorflow-great-barrier-reef')\n```",
      "votes": null
    },
    {
      "id": "1635882",
      "postDate": "01/02/2022 10:34:56",
      "content": "<p>the bug is corrected</p>\n<pre><code>bug\nprediction += '%0.8f %d %d %d %d'%(score, x, y, w, h)  \n\ncorrected:\nprediction += ' %0.8f %d %d %d %d'%(score, x, y, w, h)  \n</code></pre>",
      "rawMarkdown": "the bug is corrected\n\n```\nbug\nprediction += '%0.8f %d %d %d %d'%(score, x, y, w, h)  \n\ncorrected:\nprediction += ' %0.8f %d %d %d %d'%(score, x, y, w, h)  \n```",
      "votes": null
    },
    {
      "id": "1635990",
      "postDate": "01/02/2022 13:38:35",
      "content": "<p>you didn't write much on this competition, It is angry.</p>",
      "rawMarkdown": "you didn't write much on this competition, It is angry.",
      "votes": null
    },
    {
      "id": "1636000",
      "postDate": "01/02/2022 13:47:33",
      "content": "<p>have you tried yolox_x?   what is Its performance?  since you have the resource for training.</p>",
      "rawMarkdown": "have you tried yolox_x?   what is Its performance?  since you have the resource for training.",
      "votes": null
    },
    {
      "id": "1637150",
      "postDate": "01/03/2022 16:05:20",
      "content": "<p>Hi, was it the only bug that prevented you from submitting? My team is having issues with submission too, and we are trying to find out what the issues are:</p>\n<ol>\n<li>Did your space fix, but I think the issue is between the predictions and first prediction should not have leading whitespace</li>\n<li>Added to the system path as per comment below</li>\n</ol>\n<p>One last thing that could be the issue is having the correct number of rows, is that part of the submission error?</p>",
      "rawMarkdown": "Hi, was it the only bug that prevented you from submitting? My team is having issues with submission too, and we are trying to find out what the issues are:\n1. Did your space fix, but I think the issue is between the predictions and first prediction should not have leading whitespace\n2. Added to the system path as per comment below\n\nOne last thing that could be the issue is having the correct number of rows, is that part of the submission error?",
      "votes": null
    },
    {
      "id": "1637315",
      "postDate": "01/03/2022 18:41:55",
      "content": "<p>yes, it is the only bug. previously, there is no separation between different prediction boxes.<br>\nthe corrected space ' ' adds the required separation</p>",
      "rawMarkdown": "yes, it is the only bug. previously, there is no separation between different prediction boxes.\nthe corrected space ' ' adds the required separation",
      "votes": null
    },
    {
      "id": "1639784",
      "postDate": "01/05/2022 22:31:37",
      "content": "<p>Is the result of sample_df correct? Why it outputs single integer on the annotations column?</p>\n<pre><code>prediction:    index annotations\n0      0            \nprediction:    index annotations\n0      1            \nprediction:    index annotations\n0      2 \n</code></pre>",
      "rawMarkdown": "Is the result of sample_df correct? Why it outputs single integer on the annotations column?\n\n```\nprediction:    index annotations\n0      0            \nprediction:    index annotations\n0      1            \nprediction:    index annotations\n0      2 \n```",
      "votes": null
    },
    {
      "id": "1639785",
      "postDate": "01/05/2022 22:33:20",
      "content": "<p></p>\n<p>It seem to be solved because it was scored.<br>\nIt’s confusing (whether it solved or not is not clear at the first time). Why don’t to mark the title as [solved] for anyone first seen this discussion?</p>",
      "rawMarkdown": "~~So the original issue is solved?~~\n\nIt seem to be solved because it was scored.\nIt’s confusing (whether it solved or not is not clear at the first time). Why don’t to mark the title as [solved] for anyone first seen this discussion?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1635870,
      "author_name": "awsaf49",
      "author_url": "",
      "post_date": "01/02/2022 09:59:18",
      "content": "<p><a href=\"https://www.kaggle.com/hengck23\" target=\"_blank\">@hengck23</a> I ran into the same error a few days ago. I think this API works differently in <strong>public</strong> and <strong>private</strong>. I couldn't figure out what was wrong. Like you, my prediction and submission.csv format were all okay. After a bunch of experience here's what I found,<br>\nWe need to add <code>tensorflow-great-barrier-reef</code> to our system path even though we can import it w/o adding it in our system path.</p>\n<pre><code>import sys\nsys.path.append('../input/tensorflow-great-barrier-reef')\n</code></pre>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1635882,
      "author_name": "hengck23",
      "author_url": "",
      "post_date": "01/02/2022 10:34:56",
      "content": "<p>the bug is corrected</p>\n<pre><code>bug\nprediction += '%0.8f %d %d %d %d'%(score, x, y, w, h)  \n\ncorrected:\nprediction += ' %0.8f %d %d %d %d'%(score, x, y, w, h)  \n</code></pre>",
      "votes": null,
      "replies": [
        {
          "id": 1635990,
          "author_name": "dragonzhang",
          "author_url": "",
          "post_date": "01/02/2022 13:38:35",
          "content": "<p>you didn't write much on this competition, It is angry.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1639785,
          "author_name": "tatamikenn",
          "author_url": "",
          "post_date": "01/05/2022 22:33:20",
          "content": "<p></p>\n<p>It seem to be solved because it was scored.<br>\nIt’s confusing (whether it solved or not is not clear at the first time). Why don’t to mark the title as [solved] for anyone first seen this discussion?</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1636000,
      "author_name": "dragonzhang",
      "author_url": "",
      "post_date": "01/02/2022 13:47:33",
      "content": "<p>have you tried yolox_x?   what is Its performance?  since you have the resource for training.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1637150,
      "author_name": "hkrsmk",
      "author_url": "",
      "post_date": "01/03/2022 16:05:20",
      "content": "<p>Hi, was it the only bug that prevented you from submitting? My team is having issues with submission too, and we are trying to find out what the issues are:</p>\n<ol>\n<li>Did your space fix, but I think the issue is between the predictions and first prediction should not have leading whitespace</li>\n<li>Added to the system path as per comment below</li>\n</ol>\n<p>One last thing that could be the issue is having the correct number of rows, is that part of the submission error?</p>",
      "votes": null,
      "replies": [
        {
          "id": 1637315,
          "author_name": "hengck23",
          "author_url": "",
          "post_date": "01/03/2022 18:41:55",
          "content": "<p>yes, it is the only bug. previously, there is no separation between different prediction boxes.<br>\nthe corrected space ' ' adds the required separation</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1639784,
      "author_name": "tatamikenn",
      "author_url": "",
      "post_date": "01/05/2022 22:31:37",
      "content": "<p>Is the result of sample_df correct? Why it outputs single integer on the annotations column?</p>\n<pre><code>prediction:    index annotations\n0      0            \nprediction:    index annotations\n0      1            \nprediction:    index annotations\n0      2 \n</code></pre>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1635864": "my notebook is at https://www.kaggle.com/hengck23/my-yolox-s-1440x2560\nit has  \"Submission Scoring Error\"\n( FYI:  this is an initial code to debug my pipeline. It uses yolox-s on input 1440x2560 trained with 5 epoches)\n\nin cell [6], I use run_check_detect_one() function to check that the model and prediction function are correct.\n\ncell [5] is the submission code that uses \" iter_test = env.iter_test() \" and results in the error after running for 40 min.\ncan anyone help? thanks in advance!",
    "1635870": "hengck23 I ran into the same error a few days ago. I think this API works differently in **public** and **private**. I couldn't figure out what was wrong. Like you, my prediction and submission.csv format were all okay. After a bunch of experience here's what I found,\nWe need to add `tensorflow-great-barrier-reef` to our system path even though we can import it w/o adding it in our system path.\n```\nimport sys\nsys.path.append('../input/tensorflow-great-barrier-reef')\n```",
    "1635882": "the bug is corrected\n\n```\nbug\nprediction += '%0.8f %d %d %d %d'%(score, x, y, w, h)  \n\ncorrected:\nprediction += ' %0.8f %d %d %d %d'%(score, x, y, w, h)  \n```",
    "1635990": "you didn't write much on this competition, It is angry.",
    "1636000": "have you tried yolox_x?   what is Its performance?  since you have the resource for training.",
    "1637150": "Hi, was it the only bug that prevented you from submitting? My team is having issues with submission too, and we are trying to find out what the issues are:\n1. Did your space fix, but I think the issue is between the predictions and first prediction should not have leading whitespace\n2. Added to the system path as per comment below\n\nOne last thing that could be the issue is having the correct number of rows, is that part of the submission error?",
    "1637315": "yes, it is the only bug. previously, there is no separation between different prediction boxes.\nthe corrected space ' ' adds the required separation",
    "1639784": "Is the result of sample_df correct? Why it outputs single integer on the annotations column?\n\n```\nprediction:    index annotations\n0      0            \nprediction:    index annotations\n0      1            \nprediction:    index annotations\n0      2 \n```",
    "1639785": "~~So the original issue is solved?~~\n\nIt seem to be solved because it was scored.\nIt’s confusing (whether it solved or not is not clear at the first time). Why don’t to mark the title as [solved] for anyone first seen this discussion?"
  },
  "source": "meta"
}