{
  "id": 255263,
  "title": "My code have 'Notebook Exceeded Allowed Compute! ' error but it's running time just 2minutes",
  "url": "/competitions/mlb-player-digital-engagement-forecasting/discussion/255263",
  "author_name": "",
  "post_date": "2021-07-26T15:36:28.163971500Z",
  "votes": null,
  "comment_count": 6,
  "views": 0,
  "content": "<p>My code is composed only xgboost and its running time is just two minutes. but it is stucked and I don't know what is the problem… Someone who met like my error help me ~_~</p>",
  "messages": [
    {
      "id": "1400801",
      "postDate": "07/26/2021 15:36:28",
      "content": "<p>My code is composed only xgboost and its running time is just two minutes. but it is stucked and I don't know what is the problem… Someone who met like my error help me ~_~</p>",
      "rawMarkdown": "My code is composed only xgboost and its running time is just two minutes. but it is stucked and I don't know what is the problem... Someone who met like my error help me ~_~",
      "votes": null
    },
    {
      "id": "1400922",
      "postDate": "07/26/2021 17:57:19",
      "content": "<p>Please see <a href=\"https://www.kaggle.com/code-competition-debugging\" target=\"_blank\">https://www.kaggle.com/code-competition-debugging</a></p>\n<blockquote>\n  <p>Notebook Exceeded Allowed Compute: This indicates you have violated a code requirement constraint during the rerun. This includes limitations in the execution environment, for example requesting more RAM or disk than available, or competition constraints, such as input data source type or size limits.</p>\n</blockquote>",
      "rawMarkdown": "Please see https://www.kaggle.com/code-competition-debugging\n\n> Notebook Exceeded Allowed Compute: This indicates you have violated a code requirement constraint during the rerun. This includes limitations in the execution environment, for example requesting more RAM or disk than available, or competition constraints, such as input data source type or size limits.",
      "votes": null
    },
    {
      "id": "1400974",
      "postDate": "07/26/2021 19:09:07",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/wcukierski\" target=\"_blank\">@wcukierski</a> </p>\n<p>Notebooks that are executed without any errors in public LB will not face such an issue with the future dataset on Private LB, right?</p>",
      "rawMarkdown": "Hi @wcukierski \n\nNotebooks that are executed without any errors in public LB will not face such an issue with the future dataset on Private LB, right?",
      "votes": null
    },
    {
      "id": "1401043",
      "postDate": "07/26/2021 21:05:25",
      "content": "<p><a href=\"https://www.kaggle.com/jagadish13\" target=\"_blank\">@jagadish13</a> that's typically true, but not guaranteed to be true. New data can introduce new issues and you should attempt to code in a way that protects against surprises during stage two.</p>",
      "rawMarkdown": "jagadish13 that's typically true, but not guaranteed to be true. New data can introduce new issues and you should attempt to code in a way that protects against surprises during stage two.",
      "votes": null
    },
    {
      "id": "1401860",
      "postDate": "07/27/2021 15:52:06",
      "content": "<p>I solved this problem but the problem is in submission file not 'Notebook Exceeded Allowed Compute!'. ….</p>",
      "rawMarkdown": "I solved this problem but the problem is in submission file not 'Notebook Exceeded Allowed Compute!'. ....",
      "votes": null
    },
    {
      "id": "1401908",
      "postDate": "07/27/2021 16:25:11",
      "content": "<p>What was the issue? </p>",
      "rawMarkdown": "What was the issue?",
      "votes": null
    },
    {
      "id": "1404163",
      "postDate": "07/29/2021 15:53:43",
      "content": "<p>I had the same issue in similar circumstances. My problem was a duplication of rows for one or more players in the submission file. I had to run more than 20 submissions to figure out what the problem was. The process I followed for troubleshooting was:</p>\n<ul>\n<li>Run all my code and at the end replace my submission by sample_prediction_df filled with 0. This was important because it meant my code ran without raising exceptions and I had a way to achieve success.</li>\n<li>Test for a condition and submit sample_prediction_df filled with 0 if it is found. </li>\n<li>This enabled me to verify which conditions where present (failure=condition not found, success=condition found)</li>\n<li>Based on the past results, I'd figure out which condition to try next.</li>\n</ul>\n<p>Initially I doubted between developing a simulator or the approach above. I ended up going with the latter because I didn't have much time (I made my first submission about 1.5 weeks ago and my plan was already tight) and I hoped I could find the issue quickly. It took me a few days (having only 5 submissions per day was a drag), so it didn't go as I expected.</p>\n<p>The root of my problem was the lack of data in some of the tables (I can't remember which). It took me a while to figure it out because I coded for multiple rows for the same player/team and early in my testing, before my first submission, I realized tables could be empty and coded for it as well. Unfortunately, under certain circumstances the way I handled the absence of tables would result in a duplication of submissions for a player. For a while I couldn't understand how the duplication could be possible, but obviously there was a way. A very convoluted one.</p>\n<p>I doubt you have the exact same issue I had. I looked at a few notebooks hoping to find a tip for my issue and they all handled the lack of tables differently from me. I'd bet though that your issue has to do with the lack of input tables.</p>\n<p>Someone posted a message (or perhaps a notebook) about a simulator. By then I had already solved my issue, so I didn't take a look at it. You may want to try it.</p>\n<p>Good luck.</p>",
      "rawMarkdown": "I had the same issue in similar circumstances. My problem was a duplication of rows for one or more players in the submission file. I had to run more than 20 submissions to figure out what the problem was. The process I followed for troubleshooting was:\n- Run all my code and at the end replace my submission by sample_prediction_df filled with 0. This was important because it meant my code ran without raising exceptions and I had a way to achieve success.\n- Test for a condition and submit sample_prediction_df filled with 0 if it is found. \n- This enabled me to verify which conditions where present (failure=condition not found, success=condition found)\n- Based on the past results, I'd figure out which condition to try next.\n\nInitially I doubted between developing a simulator or the approach above. I ended up going with the latter because I didn't have much time (I made my first submission about 1.5 weeks ago and my plan was already tight) and I hoped I could find the issue quickly. It took me a few days (having only 5 submissions per day was a drag), so it didn't go as I expected.\n\nThe root of my problem was the lack of data in some of the tables (I can't remember which). It took me a while to figure it out because I coded for multiple rows for the same player/team and early in my testing, before my first submission, I realized tables could be empty and coded for it as well. Unfortunately, under certain circumstances the way I handled the absence of tables would result in a duplication of submissions for a player. For a while I couldn't understand how the duplication could be possible, but obviously there was a way. A very convoluted one.\n\nI doubt you have the exact same issue I had. I looked at a few notebooks hoping to find a tip for my issue and they all handled the lack of tables differently from me. I'd bet though that your issue has to do with the lack of input tables.\n\nSomeone posted a message (or perhaps a notebook) about a simulator. By then I had already solved my issue, so I didn't take a look at it. You may want to try it.\n\nGood luck.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1400922,
      "author_name": "wcukierski",
      "author_url": "",
      "post_date": "07/26/2021 17:57:19",
      "content": "<p>Please see <a href=\"https://www.kaggle.com/code-competition-debugging\" target=\"_blank\">https://www.kaggle.com/code-competition-debugging</a></p>\n<blockquote>\n  <p>Notebook Exceeded Allowed Compute: This indicates you have violated a code requirement constraint during the rerun. This includes limitations in the execution environment, for example requesting more RAM or disk than available, or competition constraints, such as input data source type or size limits.</p>\n</blockquote>",
      "votes": null,
      "replies": [
        {
          "id": 1400974,
          "author_name": "jagadish13",
          "author_url": "",
          "post_date": "07/26/2021 19:09:07",
          "content": "<p>Hi <a href=\"https://www.kaggle.com/wcukierski\" target=\"_blank\">@wcukierski</a> </p>\n<p>Notebooks that are executed without any errors in public LB will not face such an issue with the future dataset on Private LB, right?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1401043,
          "author_name": "wcukierski",
          "author_url": "",
          "post_date": "07/26/2021 21:05:25",
          "content": "<p><a href=\"https://www.kaggle.com/jagadish13\" target=\"_blank\">@jagadish13</a> that's typically true, but not guaranteed to be true. New data can introduce new issues and you should attempt to code in a way that protects against surprises during stage two.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1401860,
      "author_name": "developerjun",
      "author_url": "",
      "post_date": "07/27/2021 15:52:06",
      "content": "<p>I solved this problem but the problem is in submission file not 'Notebook Exceeded Allowed Compute!'. ….</p>",
      "votes": null,
      "replies": [
        {
          "id": 1401908,
          "author_name": "josephdurairaj",
          "author_url": "",
          "post_date": "07/27/2021 16:25:11",
          "content": "<p>What was the issue? </p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1404163,
      "author_name": "vialactea",
      "author_url": "",
      "post_date": "07/29/2021 15:53:43",
      "content": "<p>I had the same issue in similar circumstances. My problem was a duplication of rows for one or more players in the submission file. I had to run more than 20 submissions to figure out what the problem was. The process I followed for troubleshooting was:</p>\n<ul>\n<li>Run all my code and at the end replace my submission by sample_prediction_df filled with 0. This was important because it meant my code ran without raising exceptions and I had a way to achieve success.</li>\n<li>Test for a condition and submit sample_prediction_df filled with 0 if it is found. </li>\n<li>This enabled me to verify which conditions where present (failure=condition not found, success=condition found)</li>\n<li>Based on the past results, I'd figure out which condition to try next.</li>\n</ul>\n<p>Initially I doubted between developing a simulator or the approach above. I ended up going with the latter because I didn't have much time (I made my first submission about 1.5 weeks ago and my plan was already tight) and I hoped I could find the issue quickly. It took me a few days (having only 5 submissions per day was a drag), so it didn't go as I expected.</p>\n<p>The root of my problem was the lack of data in some of the tables (I can't remember which). It took me a while to figure it out because I coded for multiple rows for the same player/team and early in my testing, before my first submission, I realized tables could be empty and coded for it as well. Unfortunately, under certain circumstances the way I handled the absence of tables would result in a duplication of submissions for a player. For a while I couldn't understand how the duplication could be possible, but obviously there was a way. A very convoluted one.</p>\n<p>I doubt you have the exact same issue I had. I looked at a few notebooks hoping to find a tip for my issue and they all handled the lack of tables differently from me. I'd bet though that your issue has to do with the lack of input tables.</p>\n<p>Someone posted a message (or perhaps a notebook) about a simulator. By then I had already solved my issue, so I didn't take a look at it. You may want to try it.</p>\n<p>Good luck.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1400801": "My code is composed only xgboost and its running time is just two minutes. but it is stucked and I don't know what is the problem... Someone who met like my error help me ~_~",
    "1400922": "Please see https://www.kaggle.com/code-competition-debugging\n\n> Notebook Exceeded Allowed Compute: This indicates you have violated a code requirement constraint during the rerun. This includes limitations in the execution environment, for example requesting more RAM or disk than available, or competition constraints, such as input data source type or size limits.",
    "1400974": "Hi @wcukierski \n\nNotebooks that are executed without any errors in public LB will not face such an issue with the future dataset on Private LB, right?",
    "1401043": "jagadish13 that's typically true, but not guaranteed to be true. New data can introduce new issues and you should attempt to code in a way that protects against surprises during stage two.",
    "1401860": "I solved this problem but the problem is in submission file not 'Notebook Exceeded Allowed Compute!'. ....",
    "1401908": "What was the issue?",
    "1404163": "I had the same issue in similar circumstances. My problem was a duplication of rows for one or more players in the submission file. I had to run more than 20 submissions to figure out what the problem was. The process I followed for troubleshooting was:\n- Run all my code and at the end replace my submission by sample_prediction_df filled with 0. This was important because it meant my code ran without raising exceptions and I had a way to achieve success.\n- Test for a condition and submit sample_prediction_df filled with 0 if it is found. \n- This enabled me to verify which conditions where present (failure=condition not found, success=condition found)\n- Based on the past results, I'd figure out which condition to try next.\n\nInitially I doubted between developing a simulator or the approach above. I ended up going with the latter because I didn't have much time (I made my first submission about 1.5 weeks ago and my plan was already tight) and I hoped I could find the issue quickly. It took me a few days (having only 5 submissions per day was a drag), so it didn't go as I expected.\n\nThe root of my problem was the lack of data in some of the tables (I can't remember which). It took me a while to figure it out because I coded for multiple rows for the same player/team and early in my testing, before my first submission, I realized tables could be empty and coded for it as well. Unfortunately, under certain circumstances the way I handled the absence of tables would result in a duplication of submissions for a player. For a while I couldn't understand how the duplication could be possible, but obviously there was a way. A very convoluted one.\n\nI doubt you have the exact same issue I had. I looked at a few notebooks hoping to find a tip for my issue and they all handled the lack of tables differently from me. I'd bet though that your issue has to do with the lack of input tables.\n\nSomeone posted a message (or perhaps a notebook) about a simulator. By then I had already solved my issue, so I didn't take a look at it. You may want to try it.\n\nGood luck."
  },
  "source": "meta"
}