I downloaded an xgbRegressor kernel about a week ago so that I will take a look later. Today, I tried to check it out in the public kernels list but could not find it. I do not remember the owner but he/she has the to right make it private as in this one. I decided to run the kernel and share it for fairness to those who have not seen it before. You can find it here.
\n\nBy the way, it finished running and produced the result but had an error I do not understand. This is because it successfully produced the submission file that scored the same public LB=1.34 result as the original. Since there is no log I cannot determine what went wrong. If you know please let me know in the comment section below.
\n\nUPDATE
\n\nI rerun the kernel without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.
", "messages": [ { "id": "677220", "postDate": "11/19/2019 23:46:59", "content": "I downloaded an xgbRegressor kernel about a week ago so that I will take a look later. Today, I tried to check it out in the public kernels list but could not find it. I do not remember the owner but he/she has the to right make it private as in this one. I decided to run the kernel and share it for fairness to those who have not seen it before. You can find it here.
\n\nBy the way, it finished running and produced the result but had an error I do not understand. This is because it successfully produced the submission file that scored the same public LB=1.34 result as the original. Since there is no log I cannot determine what went wrong. If you know please let me know in the comment section below.
\n\nUPDATE
\n\nI rerun the kernel without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.
", "rawMarkdown": "I downloaded an xgbRegressor kernel about a week ago so that I will take a look later. Today, I tried to check it out in the public kernels list but could not find it. I do not remember the owner but he/she has the to right make it private as in this one. I decided to run the kernel and share it for fairness to those who have not seen it before. You can find it [here](https://www.kaggle.com/sheriytm/simple-xgboost-regressor-model). \n\nBy the way, it finished running and produced the result but had an error I do not understand. This is because it successfully produced the submission file that scored the same public LB=1.34 result as the original. Since there is no log I cannot determine what went wrong. If you know please let me know in the comment section below. \n\n**UPDATE**\n\nI rerun the kernel without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.", "votes": null }, { "id": "677816", "postDate": "11/20/2019 16:08:25", "content": "@sohier, can you please check if possible why my kernel ended with a red? I do not want to rerun and waste 2hrs of my GPU and end up with the same thing. Read my post above for details of my experience. Thanks.
", "rawMarkdown": "sohier, can you please check if possible why [my kernel](https://www.kaggle.com/sheriytm/simple-xgboost-regressor-model) ended with a red? I do not want to rerun and waste 2hrs of my GPU and end up with the same thing. Read my post above for details of my experience. Thanks.", "votes": null }, { "id": "679261", "postDate": "11/22/2019 14:05:24", "content": "UPDATE
\n\nI rerun the kernel without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.
", "rawMarkdown": "**UPDATE**\n\nI rerun the [kernel](https://www.kaggle.com/sheriytm/simple-xgboost-regressor-model) without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.", "votes": null }, { "id": "682522", "postDate": "11/27/2019 14:40:05", "content": "@sohier, @addisonhoward, can you please comment on my issue above? I re-ran this kernel without modification and it completed successfully. I have a private version that I ran 3X and still ends up with a red which means I cannot use its output in an ensemble. I have also wasted 6hrs of my GPU quota rerunning the private kernel. This feels frustrating with the fixed hours of GPU Kaggle is allowing at the moment.
", "rawMarkdown": "sohier, @addisonhoward, can you please comment on my issue above? I re-ran this kernel without modification and it completed successfully. I have a private version that I ran 3X and still ends up with a red which means I cannot use its output in an ensemble. I have also wasted 6hrs of my GPU quota rerunning the private kernel. This feels frustrating with the fixed hours of GPU Kaggle is allowing at the moment.", "votes": null }, { "id": "682594", "postDate": "11/27/2019 16:37:03", "content": "Looks like it was running out of disk space. I've also passed along the feedback to our engineers that more detailed error reports would be helpful.
", "rawMarkdown": "Looks like it was running out of disk space. I've also passed along the feedback to our engineers that more detailed error reports would be helpful.", "votes": null }, { "id": "682613", "postDate": "11/27/2019 17:16:26", "content": "Thanks @sohier, this is helpful and I know what to do now.
", "rawMarkdown": "Thanks @sohier, this is helpful and I know what to do now.", "votes": null }, { "id": "688257", "postDate": "12/05/2019 11:58:27", "content": "@sohier, @addisonhoward, do you have any update on this issue? I just experienced the same issue on a rerun of CPU script that does not have much to ran out of space. If it helps my script have only 15 minutes runtime.
", "rawMarkdown": "sohier, @addisonhoward, do you have any update on this issue? I just experienced the same issue on a rerun of CPU script that does not have much to ran out of space. If it helps my script have only 15 minutes runtime.", "votes": null }, { "id": "688348", "postDate": "12/05/2019 13:52:19", "content": "Could be the reason why the original author has removed this kernel.
", "rawMarkdown": "Could be the reason why the original author has removed this kernel.", "votes": null }, { "id": "688402", "postDate": "12/05/2019 14:42:21", "content": "@klloke, you may be right. It has to be broken into more than one script to run it sucessfully. That is what I did after Sohier answered me earlier. Now I am having a similar issue with a script that does not use much space so I asked the Kaggle team to check again. Since the current Kaggle kernel IDE have so many issues, I thought this may help them in debugging.
", "rawMarkdown": "klloke, you may be right. It has to be broken into more than one script to run it sucessfully. That is what I did after Sohier answered me earlier. Now I am having a similar issue with a script that does not use much space so I asked the Kaggle team to check again. Since the current Kaggle kernel IDE have so many issues, I thought this may help them in debugging.", "votes": null }, { "id": "688431", "postDate": "12/05/2019 15:24:40", "content": "I do remember Jupyter was having problem where it can't deallocate resources properly in Kaggle last time, but I'm surprised that you mention the script coding is having issue too. Could you try to insert the Except clause to print out the exception in multiple section of your script to see if it consistently died at the same place?
", "rawMarkdown": "I do remember Jupyter was having problem where it can't deallocate resources properly in Kaggle last time, but I'm surprised that you mention the script coding is having issue too. Could you try to insert the Except clause to print out the exception in multiple section of your script to see if it consistently died at the same place?", "votes": null }, { "id": "688438", "postDate": "12/05/2019 15:35:12", "content": "@klloke, sorry it is not a script but a kernel. I misspoke :-)
", "rawMarkdown": "klloke, sorry it is not a script but a kernel. I misspoke :-)", "votes": null }, { "id": "688445", "postDate": "12/05/2019 15:42:22", "content": "Try .\"New Notebook\" and then select \"Script\" to see if it helps.
", "rawMarkdown": "Try .\"New Notebook\" and then select \"Script\" to see if it helps.", "votes": null }, { "id": "688452", "postDate": "12/05/2019 15:52:12", "content": "It just finished running successfully when I tried it just now. Thanks, I will run it as a script next time so that I do not have to run it more than once.
", "rawMarkdown": "It just finished running successfully when I tried it just now. Thanks, I will run it as a script next time so that I do not have to run it more than once.", "votes": null }, { "id": "688501", "postDate": "12/05/2019 16:39:46", "content": "You're welcome. And try re-runnng it a few times again to ensure no more problem.
", "rawMarkdown": "You're welcome. And try re-runnng it a few times again to ensure no more problem.", "votes": null } ], "comments": [ { "id": 677816, "author_name": "sheriytm", "author_url": "", "post_date": "11/20/2019 16:08:25", "content": "@sohier, can you please check if possible why my kernel ended with a red? I do not want to rerun and waste 2hrs of my GPU and end up with the same thing. Read my post above for details of my experience. Thanks.
", "votes": null, "replies": [ { "id": 682522, "author_name": "sheriytm", "author_url": "", "post_date": "11/27/2019 14:40:05", "content": "@sohier, @addisonhoward, can you please comment on my issue above? I re-ran this kernel without modification and it completed successfully. I have a private version that I ran 3X and still ends up with a red which means I cannot use its output in an ensemble. I have also wasted 6hrs of my GPU quota rerunning the private kernel. This feels frustrating with the fixed hours of GPU Kaggle is allowing at the moment.
", "votes": null, "replies": [] }, { "id": 682594, "author_name": "sohier", "author_url": "", "post_date": "11/27/2019 16:37:03", "content": "Looks like it was running out of disk space. I've also passed along the feedback to our engineers that more detailed error reports would be helpful.
", "votes": null, "replies": [] }, { "id": 682613, "author_name": "sheriytm", "author_url": "", "post_date": "11/27/2019 17:16:26", "content": "Thanks @sohier, this is helpful and I know what to do now.
", "votes": null, "replies": [] }, { "id": 688257, "author_name": "sheriytm", "author_url": "", "post_date": "12/05/2019 11:58:27", "content": "@sohier, @addisonhoward, do you have any update on this issue? I just experienced the same issue on a rerun of CPU script that does not have much to ran out of space. If it helps my script have only 15 minutes runtime.
", "votes": null, "replies": [] } ] }, { "id": 679261, "author_name": "sheriytm", "author_url": "", "post_date": "11/22/2019 14:05:24", "content": "UPDATE
\n\nI rerun the kernel without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.
", "votes": null, "replies": [] }, { "id": 688348, "author_name": "klloke", "author_url": "", "post_date": "12/05/2019 13:52:19", "content": "Could be the reason why the original author has removed this kernel.
", "votes": null, "replies": [ { "id": 688402, "author_name": "sheriytm", "author_url": "", "post_date": "12/05/2019 14:42:21", "content": "@klloke, you may be right. It has to be broken into more than one script to run it sucessfully. That is what I did after Sohier answered me earlier. Now I am having a similar issue with a script that does not use much space so I asked the Kaggle team to check again. Since the current Kaggle kernel IDE have so many issues, I thought this may help them in debugging.
", "votes": null, "replies": [] }, { "id": 688431, "author_name": "klloke", "author_url": "", "post_date": "12/05/2019 15:24:40", "content": "I do remember Jupyter was having problem where it can't deallocate resources properly in Kaggle last time, but I'm surprised that you mention the script coding is having issue too. Could you try to insert the Except clause to print out the exception in multiple section of your script to see if it consistently died at the same place?
", "votes": null, "replies": [] }, { "id": 688438, "author_name": "sheriytm", "author_url": "", "post_date": "12/05/2019 15:35:12", "content": "@klloke, sorry it is not a script but a kernel. I misspoke :-)
", "votes": null, "replies": [] }, { "id": 688445, "author_name": "klloke", "author_url": "", "post_date": "12/05/2019 15:42:22", "content": "Try .\"New Notebook\" and then select \"Script\" to see if it helps.
", "votes": null, "replies": [] }, { "id": 688452, "author_name": "sheriytm", "author_url": "", "post_date": "12/05/2019 15:52:12", "content": "It just finished running successfully when I tried it just now. Thanks, I will run it as a script next time so that I do not have to run it more than once.
", "votes": null, "replies": [] }, { "id": 688501, "author_name": "klloke", "author_url": "", "post_date": "12/05/2019 16:39:46", "content": "You're welcome. And try re-runnng it a few times again to ensure no more problem.
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "677220": "I downloaded an xgbRegressor kernel about a week ago so that I will take a look later. Today, I tried to check it out in the public kernels list but could not find it. I do not remember the owner but he/she has the to right make it private as in this one. I decided to run the kernel and share it for fairness to those who have not seen it before. You can find it [here](https://www.kaggle.com/sheriytm/simple-xgboost-regressor-model). \n\nBy the way, it finished running and produced the result but had an error I do not understand. This is because it successfully produced the submission file that scored the same public LB=1.34 result as the original. Since there is no log I cannot determine what went wrong. If you know please let me know in the comment section below. \n\n**UPDATE**\n\nI rerun the kernel without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.", "677816": "sohier, can you please check if possible why [my kernel](https://www.kaggle.com/sheriytm/simple-xgboost-regressor-model) ended with a red? I do not want to rerun and waste 2hrs of my GPU and end up with the same thing. Read my post above for details of my experience. Thanks.", "679261": "**UPDATE**\n\nI rerun the [kernel](https://www.kaggle.com/sheriytm/simple-xgboost-regressor-model) without any changes and now there is no error. The problem must be Kaggle kernel issue and not related to my script when I ran version-2. Anyway, it is at least fully visible for people to see now.", "682522": "sohier, @addisonhoward, can you please comment on my issue above? I re-ran this kernel without modification and it completed successfully. I have a private version that I ran 3X and still ends up with a red which means I cannot use its output in an ensemble. I have also wasted 6hrs of my GPU quota rerunning the private kernel. This feels frustrating with the fixed hours of GPU Kaggle is allowing at the moment.", "682594": "Looks like it was running out of disk space. I've also passed along the feedback to our engineers that more detailed error reports would be helpful.", "682613": "Thanks @sohier, this is helpful and I know what to do now.", "688257": "sohier, @addisonhoward, do you have any update on this issue? I just experienced the same issue on a rerun of CPU script that does not have much to ran out of space. If it helps my script have only 15 minutes runtime.", "688348": "Could be the reason why the original author has removed this kernel.", "688402": "klloke, you may be right. It has to be broken into more than one script to run it sucessfully. That is what I did after Sohier answered me earlier. Now I am having a similar issue with a script that does not use much space so I asked the Kaggle team to check again. Since the current Kaggle kernel IDE have so many issues, I thought this may help them in debugging.", "688431": "I do remember Jupyter was having problem where it can't deallocate resources properly in Kaggle last time, but I'm surprised that you mention the script coding is having issue too. Could you try to insert the Except clause to print out the exception in multiple section of your script to see if it consistently died at the same place?", "688438": "klloke, sorry it is not a script but a kernel. I misspoke :-)", "688445": "Try .\"New Notebook\" and then select \"Script\" to see if it helps.", "688452": "It just finished running successfully when I tried it just now. Thanks, I will run it as a script next time so that I do not have to run it more than once.", "688501": "You're welcome. And try re-runnng it a few times again to ensure no more problem." }, "source": "meta" }