{
  "id": 135487,
  "title": "Scoring error after submission ... :(",
  "url": "/competitions/bengaliai-cv19/discussion/135487",
  "author_name": "",
  "post_date": "2020-03-14T04:59:34.984594400Z",
  "votes": null,
  "comment_count": 10,
  "views": 0,
  "content": "<p>I tried to submit submission.csv , but got error Submission Scoring Error.</p>\n\n<p>My submission.csv is like below</p>\n\n<p>What's wrong?</p>",
  "messages": [
    {
      "id": "771372",
      "postDate": "03/14/2020 04:59:34",
      "content": "<p>I tried to submit submission.csv , but got error Submission Scoring Error.</p>\n\n<p>My submission.csv is like below</p>\n\n<p>What's wrong?</p>",
      "rawMarkdown": "I tried to submit submission.csv , but got error Submission Scoring Error.\n\nMy submission.csv is like below\n\nWhat's wrong?",
      "votes": null
    },
    {
      "id": "771446",
      "postDate": "03/14/2020 07:28:31",
      "content": "<p>You only output prediction for sample zero... Check this out:</p>\n\n<p><a href=\"https://www.kaggle.com/corochann/bengali-seresnext-prediction-with-pytorch\">https://www.kaggle.com/corochann/bengali-seresnext-prediction-with-pytorch</a></p>",
      "rawMarkdown": "You only output prediction for sample zero... Check this out:\n\nhttps://www.kaggle.com/corochann/bengali-seresnext-prediction-with-pytorch",
      "votes": null
    },
    {
      "id": "771494",
      "postDate": "03/14/2020 08:39:13",
      "content": "<p>You have to compute the predictions in your kernel.  When you submit the test data is different than the one used in interactive or committed kernels.</p>",
      "rawMarkdown": "You have to compute the predictions in your kernel.  When you submit the test data is different than the one used in interactive or committed kernels.",
      "votes": null
    },
    {
      "id": "772513",
      "postDate": "03/15/2020 15:26:33",
      "content": "<p>Thanks!\nI uploaded my submission.csv computed in offline, but should I compute on own kernel?</p>",
      "rawMarkdown": "Thanks!\nI uploaded my submission.csv computed in offline, but should I compute on own kernel?",
      "votes": null
    },
    {
      "id": "772533",
      "postDate": "03/15/2020 15:40:13",
      "content": "<p>You need to commit your kernel first, which will generate predictions online, then you commit the output.  Please check this out: <a href=\"https://www.kaggle.com/c/bengaliai-cv19/submit\">https://www.kaggle.com/c/bengaliai-cv19/submit</a></p>",
      "rawMarkdown": "You need to commit your kernel first, which will generate predictions online, then you commit the output.  Please check this out: https://www.kaggle.com/c/bengaliai-cv19/submit",
      "votes": null
    },
    {
      "id": "772626",
      "postDate": "03/15/2020 17:36:40",
      "content": "<p>I cannot ... got scoring error</p>",
      "rawMarkdown": "I cannot ... got scoring error",
      "votes": null
    },
    {
      "id": "772662",
      "postDate": "03/15/2020 18:30:42",
      "content": "<p>Did you run your kernel successfully?  If you did, it might be the submission file error. The result file name should be submission.csv,  and the file row numbers should be adjustable for the test set length. Did you read the notebook's requirements? \n<a href=\"https://www.kaggle.com/c/bengaliai-cv19/overview/notebooks-requirements\">https://www.kaggle.com/c/bengaliai-cv19/overview/notebooks-requirements</a></p>",
      "rawMarkdown": "Did you run your kernel successfully?  If you did, it might be the submission file error. The result file name should be submission.csv,  and the file row numbers should be adjustable for the test set length. Did you read the notebook's requirements? \nhttps://www.kaggle.com/c/bengaliai-cv19/overview/notebooks-requirements",
      "votes": null
    },
    {
      "id": "772786",
      "postDate": "03/15/2020 22:52:42",
      "content": "<p>Yes, but submission doesn't run.\nI changed notebook to run inferencce model and make submission.csv.\nThis change causes long time submission after \"submit to competition\", is this usual? What happened?</p>",
      "rawMarkdown": "Yes, but submission doesn't run.\nI changed notebook to run inferencce model and make submission.csv.\nThis change causes long time submission after \"submit to competition\", is this usual? What happened?",
      "votes": null
    },
    {
      "id": "772794",
      "postDate": "03/15/2020 23:04:47",
      "content": "<p>I think so. After you submission, the kernel will be rerun with the test data set on the kaggle platform. If you turn on your gpu, a single model usually causes 15 mins, the whole scoring process takes roughly (15 mins * models number you used).  If you don't use gpu to accelerate, the process will take much longer. I don't use cpu for submission , so I don't know how long it will take. Good luck!</p>",
      "rawMarkdown": "I think so. After you submission, the kernel will be rerun with the test data set on the kaggle platform. If you turn on your gpu, a single model usually causes 15 mins, the whole scoring process takes roughly (15 mins * models number you used).  If you don't use gpu to accelerate, the process will take much longer. I don't use cpu for submission , so I don't know how long it will take. Good luck!",
      "votes": null
    },
    {
      "id": "775146",
      "postDate": "03/16/2020 11:05:02",
      "content": "<p>Thank you for your detail description.\nI turned on GPU, then my submission is notebook time out.\nI gave up this competition :(</p>",
      "rawMarkdown": "Thank you for your detail description.\nI turned on GPU, then my submission is notebook time out.\nI gave up this competition :(",
      "votes": null
    },
    {
      "id": "775508",
      "postDate": "03/16/2020 18:47:33",
      "content": "<p>Sorry to hear that. Good luck next time!</p>",
      "rawMarkdown": "Sorry to hear that. Good luck next time!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 771446,
      "author_name": "yuanlin08",
      "author_url": "",
      "post_date": "03/14/2020 07:28:31",
      "content": "<p>You only output prediction for sample zero... Check this out:</p>\n\n<p><a href=\"https://www.kaggle.com/corochann/bengali-seresnext-prediction-with-pytorch\">https://www.kaggle.com/corochann/bengali-seresnext-prediction-with-pytorch</a></p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 771494,
      "author_name": "cpmpml",
      "author_url": "",
      "post_date": "03/14/2020 08:39:13",
      "content": "<p>You have to compute the predictions in your kernel.  When you submit the test data is different than the one used in interactive or committed kernels.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 772513,
      "author_name": "yoshito",
      "author_url": "",
      "post_date": "03/15/2020 15:26:33",
      "content": "<p>Thanks!\nI uploaded my submission.csv computed in offline, but should I compute on own kernel?</p>",
      "votes": null,
      "replies": [
        {
          "id": 772533,
          "author_name": "yuanlin08",
          "author_url": "",
          "post_date": "03/15/2020 15:40:13",
          "content": "<p>You need to commit your kernel first, which will generate predictions online, then you commit the output.  Please check this out: <a href=\"https://www.kaggle.com/c/bengaliai-cv19/submit\">https://www.kaggle.com/c/bengaliai-cv19/submit</a></p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 772626,
          "author_name": "yoshito",
          "author_url": "",
          "post_date": "03/15/2020 17:36:40",
          "content": "<p>I cannot ... got scoring error</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 772662,
          "author_name": "yuanlin08",
          "author_url": "",
          "post_date": "03/15/2020 18:30:42",
          "content": "<p>Did you run your kernel successfully?  If you did, it might be the submission file error. The result file name should be submission.csv,  and the file row numbers should be adjustable for the test set length. Did you read the notebook's requirements? \n<a href=\"https://www.kaggle.com/c/bengaliai-cv19/overview/notebooks-requirements\">https://www.kaggle.com/c/bengaliai-cv19/overview/notebooks-requirements</a></p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 772786,
          "author_name": "yoshito",
          "author_url": "",
          "post_date": "03/15/2020 22:52:42",
          "content": "<p>Yes, but submission doesn't run.\nI changed notebook to run inferencce model and make submission.csv.\nThis change causes long time submission after \"submit to competition\", is this usual? What happened?</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 772794,
          "author_name": "yuanlin08",
          "author_url": "",
          "post_date": "03/15/2020 23:04:47",
          "content": "<p>I think so. After you submission, the kernel will be rerun with the test data set on the kaggle platform. If you turn on your gpu, a single model usually causes 15 mins, the whole scoring process takes roughly (15 mins * models number you used).  If you don't use gpu to accelerate, the process will take much longer. I don't use cpu for submission , so I don't know how long it will take. Good luck!</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 775146,
          "author_name": "yoshito",
          "author_url": "",
          "post_date": "03/16/2020 11:05:02",
          "content": "<p>Thank you for your detail description.\nI turned on GPU, then my submission is notebook time out.\nI gave up this competition :(</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 775508,
          "author_name": "yuanlin08",
          "author_url": "",
          "post_date": "03/16/2020 18:47:33",
          "content": "<p>Sorry to hear that. Good luck next time!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "771372": "I tried to submit submission.csv , but got error Submission Scoring Error.\n\nMy submission.csv is like below\n\nWhat's wrong?",
    "771446": "You only output prediction for sample zero... Check this out:\n\nhttps://www.kaggle.com/corochann/bengali-seresnext-prediction-with-pytorch",
    "771494": "You have to compute the predictions in your kernel.  When you submit the test data is different than the one used in interactive or committed kernels.",
    "772513": "Thanks!\nI uploaded my submission.csv computed in offline, but should I compute on own kernel?",
    "772533": "You need to commit your kernel first, which will generate predictions online, then you commit the output.  Please check this out: https://www.kaggle.com/c/bengaliai-cv19/submit",
    "772626": "I cannot ... got scoring error",
    "772662": "Did you run your kernel successfully?  If you did, it might be the submission file error. The result file name should be submission.csv,  and the file row numbers should be adjustable for the test set length. Did you read the notebook's requirements? \nhttps://www.kaggle.com/c/bengaliai-cv19/overview/notebooks-requirements",
    "772786": "Yes, but submission doesn't run.\nI changed notebook to run inferencce model and make submission.csv.\nThis change causes long time submission after \"submit to competition\", is this usual? What happened?",
    "772794": "I think so. After you submission, the kernel will be rerun with the test data set on the kaggle platform. If you turn on your gpu, a single model usually causes 15 mins, the whole scoring process takes roughly (15 mins * models number you used).  If you don't use gpu to accelerate, the process will take much longer. I don't use cpu for submission , so I don't know how long it will take. Good luck!",
    "775146": "Thank you for your detail description.\nI turned on GPU, then my submission is notebook time out.\nI gave up this competition :(",
    "775508": "Sorry to hear that. Good luck next time!"
  },
  "source": "meta"
}