![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F367007%2Fe47f483908fdd3ddaba9f23d2c464cfc%2FScreenshot%202019-08-21%20at%2014.23.03.png?generation=1566390302936852&alt=media\")
I am trying to make a submission (my first time to use kernel submissions), I got this error. Can anyone help here? My kernel uploads a model, reads in the test data and then make predictions, and finally stores them in submission.csv:
\n\n
I am trying to make a submission (my first time to use kernel submissions), I got this error. Can anyone help here? My kernel uploads a model, reads in the test data and then make predictions, and finally stores them in submission.csv:
\n\n
you should follow a successful kernel and edit their code to make your submission
", "rawMarkdown": "you should follow a successful kernel and edit their code to make your submission", "votes": null }, { "id": "605677", "postDate": "08/22/2019 16:03:14", "content": "I did use one that ran successfully, mine also ran successfully, I also made sure that I read the sample_submission file and populate it with predictions, and I also made sure that the order of rows is as in the submission file, but I still have this error!!!
", "rawMarkdown": "I did use one that ran successfully, mine also ran successfully, I also made sure that I read the sample_submission file and populate it with predictions, and I also made sure that the order of rows is as in the submission file, but I still have this error!!!", "votes": null }, { "id": "605751", "postDate": "08/22/2019 17:38:16", "content": "It is not you but Ksggle's fault. It has been reported here and in other threads.
", "rawMarkdown": "It is not you but Ksggle's fault. It has been reported [here](https://www.kaggle.com/c/aptos2019-blindness-detection/discussion/102998#latest-593633) and in other threads.", "votes": null }, { "id": "606279", "postDate": "08/23/2019 11:41:28", "content": "Hi,
\n\nDo you have by any chance a working code (lets say with very low accuracy) that works fine to submit on the competition?
\n\nI do the following:\n1) Run the code\n2) Commit\n3) Click on Submit to competition\n4) Get Submission CSV error
\n\nMy code runs absolutely fine, but I am not able to find any reason
", "rawMarkdown": "Hi, \n\nDo you have by any chance a working code (lets say with very low accuracy) that works fine to submit on the competition?\n\nI do the following:\n1) Run the code\n2) Commit\n3) Click on Submit to competition\n4) Get Submission CSV error\n\nMy code runs absolutely fine, but I am not able to find any reason", "votes": null }, { "id": "606629", "postDate": "08/23/2019 20:03:43", "content": "Check the completed notebook. When a notebook is done running, you can see the output of each cell when reviewing the committed version. I would check there first for errors.
", "rawMarkdown": "Check the completed notebook. When a notebook is done running, you can see the output of each cell when reviewing the committed version. I would check there first for errors.", "votes": null }, { "id": "606634", "postDate": "08/23/2019 20:12:14", "content": "I don't have much of a cells, because I am doing most of the processes offline, and then I upload the model, read the test data, do predictions save them. Here is the message I get from my last cell where the submission is completed, and then you can see in the attached screen shot how I am saving it. I am reading the sample submission to make sure it is all in order: \nam I doing anything wrong? \n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F367007%2F9a23861f2836bf54fea2d7a50b53a686%2FScreenshot%202019-08-23%20at%2022.09.49.png?generation=1566591018553714&alt=media\")
I doubt the issue has to do with your submission file -- at least, we wouldn't know that until you submit and you get a score of 0.0000.
\n\n\"Kernel Threw Exception\" refers to the kernel crashing when trying to run on the test data. If your kernel is running fine on the public test set, then there is likely something wrong with how you may be processing those images such that it's crashing on the private test set.
\n\nWithout knowing more about how you are processing the test images, I wouldn't be able to say why.
\n\nFor the sake of completeness: In Synchronous Kernels-only Competitions, the public test set is substituted for the private data set at run-time. Any processing you do on the public test set must be robust enough to run on an unseen test set without errors. It's a bit of a guessing game.
", "rawMarkdown": "I doubt the issue has to do with your submission file -- at least, we wouldn't know that until you submit and you get a score of 0.0000. \n\n\"Kernel Threw Exception\" refers to the kernel crashing when trying to run on the test data. If your kernel is running fine on the public test set, then there is likely something wrong with how you may be processing those images such that it's crashing on the private test set. \n\nWithout knowing more about how you are processing the test images, I wouldn't be able to say why. \n\nFor the sake of completeness: In Synchronous Kernels-only Competitions, the public test set is substituted for the private data set at run-time. Any processing you do on the public test set must be robust enough to run on an unseen test set without errors. It's a bit of a guessing game.", "votes": null }, { "id": "606683", "postDate": "08/23/2019 22:56:11", "content": "This is a good kernel that you can follow in order to achieve successful submission\nhttps://www.kaggle.com/drhabib/starter-kernel-for-0-79
\n\nThe main instructions to follow are:\n1. Upload your weight if you have trained in your local machine and do not upload the predicted csv and submit by any chance. You should create a submission.csv from the prediction of your model.\n2. If you have trained a model in kernel just follow the above kernel to make successful submissions.
", "rawMarkdown": "This is a good kernel that you can follow in order to achieve successful submission\nhttps://www.kaggle.com/drhabib/starter-kernel-for-0-79\n\nThe main instructions to follow are:\n1. Upload your weight if you have trained in your local machine and do not upload the predicted csv and submit by any chance. You should create a submission.csv from the prediction of your model.\n2. If you have trained a model in kernel just follow the above kernel to make successful submissions.", "votes": null } ], "comments": [ { "id": 604561, "author_name": "sabbiracoustic1006", "author_url": "", "post_date": "08/21/2019 14:16:12", "content": "you should follow a successful kernel and edit their code to make your submission
", "votes": null, "replies": [ { "id": 606279, "author_name": "datachampion", "author_url": "", "post_date": "08/23/2019 11:41:28", "content": "Hi,
\n\nDo you have by any chance a working code (lets say with very low accuracy) that works fine to submit on the competition?
\n\nI do the following:\n1) Run the code\n2) Commit\n3) Click on Submit to competition\n4) Get Submission CSV error
\n\nMy code runs absolutely fine, but I am not able to find any reason
", "votes": null, "replies": [] }, { "id": 606683, "author_name": "sabbiracoustic1006", "author_url": "", "post_date": "08/23/2019 22:56:11", "content": "This is a good kernel that you can follow in order to achieve successful submission\nhttps://www.kaggle.com/drhabib/starter-kernel-for-0-79
\n\nThe main instructions to follow are:\n1. Upload your weight if you have trained in your local machine and do not upload the predicted csv and submit by any chance. You should create a submission.csv from the prediction of your model.\n2. If you have trained a model in kernel just follow the above kernel to make successful submissions.
", "votes": null, "replies": [] } ] }, { "id": 605677, "author_name": "owise83", "author_url": "", "post_date": "08/22/2019 16:03:14", "content": "I did use one that ran successfully, mine also ran successfully, I also made sure that I read the sample_submission file and populate it with predictions, and I also made sure that the order of rows is as in the submission file, but I still have this error!!!
", "votes": null, "replies": [] }, { "id": 605751, "author_name": "sheriytm", "author_url": "", "post_date": "08/22/2019 17:38:16", "content": "It is not you but Ksggle's fault. It has been reported here and in other threads.
", "votes": null, "replies": [] }, { "id": 606629, "author_name": "amqdnguyen", "author_url": "", "post_date": "08/23/2019 20:03:43", "content": "Check the completed notebook. When a notebook is done running, you can see the output of each cell when reviewing the committed version. I would check there first for errors.
", "votes": null, "replies": [ { "id": 606634, "author_name": "owise83", "author_url": "", "post_date": "08/23/2019 20:12:14", "content": "I don't have much of a cells, because I am doing most of the processes offline, and then I upload the model, read the test data, do predictions save them. Here is the message I get from my last cell where the submission is completed, and then you can see in the attached screen shot how I am saving it. I am reading the sample submission to make sure it is all in order: \nam I doing anything wrong? \n
I doubt the issue has to do with your submission file -- at least, we wouldn't know that until you submit and you get a score of 0.0000.
\n\n\"Kernel Threw Exception\" refers to the kernel crashing when trying to run on the test data. If your kernel is running fine on the public test set, then there is likely something wrong with how you may be processing those images such that it's crashing on the private test set.
\n\nWithout knowing more about how you are processing the test images, I wouldn't be able to say why.
\n\nFor the sake of completeness: In Synchronous Kernels-only Competitions, the public test set is substituted for the private data set at run-time. Any processing you do on the public test set must be robust enough to run on an unseen test set without errors. It's a bit of a guessing game.
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "604467": "I am trying to make a submission (my first time to use kernel submissions), I got this error. Can anyone help here? My kernel uploads a model, reads in the test data and then make predictions, and finally stores them in submission.csv:\n\n\n", "604561": "you should follow a successful kernel and edit their code to make your submission", "605677": "I did use one that ran successfully, mine also ran successfully, I also made sure that I read the sample_submission file and populate it with predictions, and I also made sure that the order of rows is as in the submission file, but I still have this error!!!", "605751": "It is not you but Ksggle's fault. It has been reported [here](https://www.kaggle.com/c/aptos2019-blindness-detection/discussion/102998#latest-593633) and in other threads.", "606279": "Hi, \n\nDo you have by any chance a working code (lets say with very low accuracy) that works fine to submit on the competition?\n\nI do the following:\n1) Run the code\n2) Commit\n3) Click on Submit to competition\n4) Get Submission CSV error\n\nMy code runs absolutely fine, but I am not able to find any reason", "606629": "Check the completed notebook. When a notebook is done running, you can see the output of each cell when reviewing the committed version. I would check there first for errors.", "606634": "I don't have much of a cells, because I am doing most of the processes offline, and then I upload the model, read the test data, do predictions save them. Here is the message I get from my last cell where the submission is completed, and then you can see in the attached screen shot how I am saving it. I am reading the sample submission to make sure it is all in order: \nam I doing anything wrong? \n", "606643": "I doubt the issue has to do with your submission file -- at least, we wouldn't know that until you submit and you get a score of 0.0000. \n\n\"Kernel Threw Exception\" refers to the kernel crashing when trying to run on the test data. If your kernel is running fine on the public test set, then there is likely something wrong with how you may be processing those images such that it's crashing on the private test set. \n\nWithout knowing more about how you are processing the test images, I wouldn't be able to say why. \n\nFor the sake of completeness: In Synchronous Kernels-only Competitions, the public test set is substituted for the private data set at run-time. Any processing you do on the public test set must be robust enough to run on an unseen test set without errors. It's a bit of a guessing game.", "606683": "This is a good kernel that you can follow in order to achieve successful submission\nhttps://www.kaggle.com/drhabib/starter-kernel-for-0-79\n\nThe main instructions to follow are:\n1. Upload your weight if you have trained in your local machine and do not upload the predicted csv and submit by any chance. You should create a submission.csv from the prediction of your model.\n2. If you have trained a model in kernel just follow the above kernel to make successful submissions." }, "source": "meta" }