{
  "id": 126922,
  "title": " help!!! submit-to-competition always error Notebook Exceeded Allowed Compute",
  "url": "/competitions/bengaliai-cv19/discussion/126922",
  "author_name": "",
  "post_date": "2020-01-21T07:00:03.077750100Z",
  "votes": 1,
  "comment_count": 8,
  "views": 0,
  "content": "<p>mykaggle_bengali_minist2(version 4/4)\n2 minutes ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Running\nmykaggle_bengali_minist2(version 4/4)\n25 minutes ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 3/4)\n21 hours ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 2/4)\na day ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 1/4)</p>",
  "messages": [
    {
      "id": "724462",
      "postDate": "01/21/2020 07:00:03",
      "content": "<p>mykaggle_bengali_minist2(version 4/4)\n2 minutes ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Running\nmykaggle_bengali_minist2(version 4/4)\n25 minutes ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 3/4)\n21 hours ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 2/4)\na day ago by Zhiquantime</p>\n\n<p>From Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 1/4)</p>",
      "rawMarkdown": "mykaggle_bengali_minist2(version 4/4)\n2 minutes ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Running\nmykaggle_bengali_minist2(version 4/4)\n25 minutes ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 3/4)\n21 hours ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 2/4)\na day ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 1/4)",
      "votes": null
    },
    {
      "id": "724587",
      "postDate": "01/21/2020 09:17:27",
      "content": "<p>crazy! help!</p>",
      "rawMarkdown": "crazy! help!",
      "votes": null
    },
    {
      "id": "725730",
      "postDate": "01/22/2020 12:51:58",
      "content": "<p>+1 I have try reduce memory cost(both cpu and cuda). local test: 2G memory cost 1G cuda memory cost .running time just a few minutes. what cost the error?</p>",
      "rawMarkdown": "1 I have try reduce memory cost(both cpu and cuda). local test: 2G memory cost 1G cuda memory cost .running time just a few minutes. what cost the error?",
      "votes": null
    },
    {
      "id": "725871",
      "postDate": "01/22/2020 15:02:35",
      "content": "<p><a href=\"/zhiquantime\">@zhiquantime</a> Need more explanation about the model and library you are using!! for example if you are doing deep learning using Keras you can save the model as .h5 file and then use it while submitting !!</p>\n\n<ol>\n<li>Avoid training in the Kernal you are submitting -&gt; this will save you a great amount of time</li>\n<li>Simplify the model (Unwanted complexity is a waste of time)</li>\n<li>Check were you are taking more time using <a href=\"https://docs.python.org/2/library/timeit.html\">python timit</a></li>\n</ol>\n\n<p>Give me more info I can help you out (@ me so that I can get, notification immediately)\nIf my post helped you please upvote, thanks!!</p>",
      "rawMarkdown": "zhiquantime Need more explanation about the model and library you are using!! for example if you are doing deep learning using Keras you can save the model as .h5 file and then use it while submitting !!\n\n1. Avoid training in the Kernal you are submitting -&gt; this will save you a great amount of time\n2. Simplify the model (Unwanted complexity is a waste of time)\n3. Check were you are taking more time using [python timit](https://docs.python.org/2/library/timeit.html)\n\nGive me more info I can help you out (@ me so that I can get, notification immediately)\nIf my post helped you please upvote, thanks!!",
      "votes": null
    },
    {
      "id": "726385",
      "postDate": "01/23/2020 01:42:36",
      "content": "<p><a href=\"/chekoduadarsh\">@chekoduadarsh</a> , Mr.Shadow,thank you very much👍 👍 ！！I will try your suggestions!</p>",
      "rawMarkdown": "chekoduadarsh , Mr.Shadow,thank you very much👍 👍 ！！I will try your suggestions!",
      "votes": null
    },
    {
      "id": "726423",
      "postDate": "01/23/2020 02:17:00",
      "content": "<p><a href=\"/chekoduadarsh\">@chekoduadarsh</a> ,Mr.Shadow,can you explain how to save mode as .h5 in notebook and then read the model in comitting??  I could not get usefull infomation in google..</p>",
      "rawMarkdown": "chekoduadarsh ,Mr.Shadow,can you explain how to save mode as .h5 in notebook and then read the model in comitting??  I could not get usefull infomation in google..",
      "votes": null
    },
    {
      "id": "727155",
      "postDate": "01/23/2020 13:55:29",
      "content": "<p><a href=\"/zhiquantime\">@zhiquantime</a> \n<a href=\"https://machinelearningmastery.com/save-load-keras-deep-learning-models/\">https://machinelearningmastery.com/save-load-keras-deep-learning-models/</a>\n<a href=\"https://www.tensorflow.org/guide/keras/save_and_serialize\">https://www.tensorflow.org/guide/keras/save_and_serialize</a>\n<a href=\"https://stackoverflow.com/questions/42763094/how-to-save-final-model-using-keras\">https://stackoverflow.com/questions/42763094/how-to-save-final-model-using-keras</a></p>",
      "rawMarkdown": "zhiquantime \nhttps://machinelearningmastery.com/save-load-keras-deep-learning-models/\nhttps://www.tensorflow.org/guide/keras/save_and_serialize\nhttps://stackoverflow.com/questions/42763094/how-to-save-final-model-using-keras",
      "votes": null
    },
    {
      "id": "736283",
      "postDate": "02/04/2020 02:46:52",
      "content": "<p>Mr.Shadow，I have already done:\n(1) train model in <a href=\"https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist2\">mykaggle-bengali-minist2</a>\n(2)predict test's result  in <a href=\"https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist3\">mykaggle-bengali-minist3</a> by link the model of  mykaggle-bengali-minist2 </p>\n\n<p>but the same problem \"Notebook Exceeded Allowed Compute\" in my submission ---- mykaggle-bengali-minist3。The GPU  full-time of  mykaggle-bengali-minist3 is &lt;=30min</p>",
      "rawMarkdown": "Mr.Shadow，I have already done:\n(1) train model in [mykaggle-bengali-minist2](https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist2)\n(2)predict test's result  in [mykaggle-bengali-minist3](https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist3) by link the model of  mykaggle-bengali-minist2 \n\nbut the same problem \"Notebook Exceeded Allowed Compute\" in my submission ---- mykaggle-bengali-minist3。The GPU  full-time of  mykaggle-bengali-minist3 is &lt;=30min",
      "votes": null
    },
    {
      "id": "744282",
      "postDate": "02/12/2020 17:40:50",
      "content": "<p>I've been having huge issues with this and I couldn't even submit 1's.  So far I found that when I go to reshape the data, that seems to be the breaking point for me.  I created a public notebook related to my findings.\n<a href=\"https://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help\">https://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help</a></p>",
      "rawMarkdown": "I've been having huge issues with this and I couldn't even submit 1's.  So far I found that when I go to reshape the data, that seems to be the breaking point for me.  I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 724587,
      "author_name": "zhiquantime",
      "author_url": "",
      "post_date": "01/21/2020 09:17:27",
      "content": "<p>crazy! help!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 725730,
      "author_name": "a457435687",
      "author_url": "",
      "post_date": "01/22/2020 12:51:58",
      "content": "<p>+1 I have try reduce memory cost(both cpu and cuda). local test: 2G memory cost 1G cuda memory cost .running time just a few minutes. what cost the error?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 725871,
      "author_name": "chekoduadarsh",
      "author_url": "",
      "post_date": "01/22/2020 15:02:35",
      "content": "<p><a href=\"/zhiquantime\">@zhiquantime</a> Need more explanation about the model and library you are using!! for example if you are doing deep learning using Keras you can save the model as .h5 file and then use it while submitting !!</p>\n\n<ol>\n<li>Avoid training in the Kernal you are submitting -&gt; this will save you a great amount of time</li>\n<li>Simplify the model (Unwanted complexity is a waste of time)</li>\n<li>Check were you are taking more time using <a href=\"https://docs.python.org/2/library/timeit.html\">python timit</a></li>\n</ol>\n\n<p>Give me more info I can help you out (@ me so that I can get, notification immediately)\nIf my post helped you please upvote, thanks!!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 726385,
      "author_name": "zhiquantime",
      "author_url": "",
      "post_date": "01/23/2020 01:42:36",
      "content": "<p><a href=\"/chekoduadarsh\">@chekoduadarsh</a> , Mr.Shadow,thank you very much👍 👍 ！！I will try your suggestions!</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 726423,
      "author_name": "zhiquantime",
      "author_url": "",
      "post_date": "01/23/2020 02:17:00",
      "content": "<p><a href=\"/chekoduadarsh\">@chekoduadarsh</a> ,Mr.Shadow,can you explain how to save mode as .h5 in notebook and then read the model in comitting??  I could not get usefull infomation in google..</p>",
      "votes": null,
      "replies": [
        {
          "id": 727155,
          "author_name": "chekoduadarsh",
          "author_url": "",
          "post_date": "01/23/2020 13:55:29",
          "content": "<p><a href=\"/zhiquantime\">@zhiquantime</a> \n<a href=\"https://machinelearningmastery.com/save-load-keras-deep-learning-models/\">https://machinelearningmastery.com/save-load-keras-deep-learning-models/</a>\n<a href=\"https://www.tensorflow.org/guide/keras/save_and_serialize\">https://www.tensorflow.org/guide/keras/save_and_serialize</a>\n<a href=\"https://stackoverflow.com/questions/42763094/how-to-save-final-model-using-keras\">https://stackoverflow.com/questions/42763094/how-to-save-final-model-using-keras</a></p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 736283,
          "author_name": "zhiquantime",
          "author_url": "",
          "post_date": "02/04/2020 02:46:52",
          "content": "<p>Mr.Shadow，I have already done:\n(1) train model in <a href=\"https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist2\">mykaggle-bengali-minist2</a>\n(2)predict test's result  in <a href=\"https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist3\">mykaggle-bengali-minist3</a> by link the model of  mykaggle-bengali-minist2 </p>\n\n<p>but the same problem \"Notebook Exceeded Allowed Compute\" in my submission ---- mykaggle-bengali-minist3。The GPU  full-time of  mykaggle-bengali-minist3 is &lt;=30min</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 744282,
      "author_name": "yeayates21",
      "author_url": "",
      "post_date": "02/12/2020 17:40:50",
      "content": "<p>I've been having huge issues with this and I couldn't even submit 1's.  So far I found that when I go to reshape the data, that seems to be the breaking point for me.  I created a public notebook related to my findings.\n<a href=\"https://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help\">https://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "724462": "mykaggle_bengali_minist2(version 4/4)\n2 minutes ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Running\nmykaggle_bengali_minist2(version 4/4)\n25 minutes ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 3/4)\n21 hours ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 2/4)\na day ago by Zhiquantime\n\nFrom Kernel [mykaggle_bengali_minist2]\nNotebook Exceeded Allowed Compute\nError\nmykaggle_bengali_minist2(version 1/4)",
    "724587": "crazy! help!",
    "725730": "1 I have try reduce memory cost(both cpu and cuda). local test: 2G memory cost 1G cuda memory cost .running time just a few minutes. what cost the error?",
    "725871": "zhiquantime Need more explanation about the model and library you are using!! for example if you are doing deep learning using Keras you can save the model as .h5 file and then use it while submitting !!\n\n1. Avoid training in the Kernal you are submitting -&gt; this will save you a great amount of time\n2. Simplify the model (Unwanted complexity is a waste of time)\n3. Check were you are taking more time using [python timit](https://docs.python.org/2/library/timeit.html)\n\nGive me more info I can help you out (@ me so that I can get, notification immediately)\nIf my post helped you please upvote, thanks!!",
    "726385": "chekoduadarsh , Mr.Shadow,thank you very much👍 👍 ！！I will try your suggestions!",
    "726423": "chekoduadarsh ,Mr.Shadow,can you explain how to save mode as .h5 in notebook and then read the model in comitting??  I could not get usefull infomation in google..",
    "727155": "zhiquantime \nhttps://machinelearningmastery.com/save-load-keras-deep-learning-models/\nhttps://www.tensorflow.org/guide/keras/save_and_serialize\nhttps://stackoverflow.com/questions/42763094/how-to-save-final-model-using-keras",
    "736283": "Mr.Shadow，I have already done:\n(1) train model in [mykaggle-bengali-minist2](https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist2)\n(2)predict test's result  in [mykaggle-bengali-minist3](https://www.kaggle.com/zhiquantime/mykaggle-bengali-minist3) by link the model of  mykaggle-bengali-minist2 \n\nbut the same problem \"Notebook Exceeded Allowed Compute\" in my submission ---- mykaggle-bengali-minist3。The GPU  full-time of  mykaggle-bengali-minist3 is &lt;=30min",
    "744282": "I've been having huge issues with this and I couldn't even submit 1's.  So far I found that when I go to reshape the data, that seems to be the breaking point for me.  I created a public notebook related to my findings.\nhttps://www.kaggle.com/yeayates21/cant-submit-1s-whats-wrong-please-help"
  },
  "source": "meta"
}