Hello all.
\nFirst, thanks for the team hosting this challenge, it's quite interesting.
\nI'd like to ask others participating in this challenge if you can check out if there's anything wrong about my submissions, as I keep having a Submission Scoring Error, and I can't see what's wrong with the output (disregarding the accuracy of results).
\n![\"submission](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F11662325%2F1395a63cbfbeef84f5f94d58481f5a60%2Foutput.jpg?generation=1678983450390672&alt=media\")
\nHere is my kernel: https://www.kaggle.com/code/sofiamatias/icecube-predictions
Thanks a lot in advance and good luck to all!
\nSofia
", "messages": [ { "id": 2188949, "postDate": "2023-03-20T05:44:59.390Z", "content": "Try the following
\nsubmission_df.to_csv('/kaggle/working/submission.csv', index=False)
\nIt seems the directory is missing in your notebook.
As noted by Alex Z your notebook is not looking at all of the test batches. This notebook using catboost would be similiar to yours and you can see how the unknown number of batches that will be in the hidden test can be processed.
\n(https://www.kaggle.com/code/astrung/pcaembed-gpu-catboost-optuna-approach)
\nThat notebook was based on a shared xgboost and it looks like the same code was usable for both a catboost and an xgb approach. Think this one is the origin code for the submission section.
\nIceCube🧊: Neutrino🎆 baseline XGBoost🚀Regression
", "rawMarkdown": "As noted by [Alex Z ](https://www.kaggle.com/alexz0)your notebook is not looking at all of the test batches. This notebook using catboost would be similiar to yours and you can see how the unknown number of batches that will be in the hidden test can be processed.\n\n(https://www.kaggle.com/code/astrung/pcaembed-gpu-catboost-optuna-approach)\n\nThat notebook was based on a shared xgboost and it looks like the same code was usable for both a catboost and an xgb approach. Think this one is the origin code for the submission section.\n\n[IceCube🧊: Neutrino🎆 baseline XGBoost🚀Regression](uhttps://www.kaggle.com/code/jirkaborovec/icecube-neutrino-baseline-xgboost-regressionrl)\n", "replies": [ { "id": 2185705, "postDate": "2023-03-17T08:37:15.220Z", "content": "Thanks, will look into it.
", "rawMarkdown": "Thanks, will look into it." } ] }, { "id": 2185052, "postDate": "2023-03-16T19:16:13.837Z", "content": "Batch 661 is a dummy sample with only three events, the actual test dataset is hidden. When doing inference you should read all the test batches from test/batch_[n].parquet
", "rawMarkdown": "Batch 661 is a dummy sample with only three events, the actual test dataset is hidden. When doing inference you should read all the test batches from test/batch_[n].parquet", "replies": [ { "id": 2185087, "postDate": "2023-03-16T19:45:12.780Z", "content": "Aha, that's it! I'll look into it, thanks!!
", "rawMarkdown": "Aha, that's it! I'll look into it, thanks!!" } ] }, { "id": 2184867, "postDate": "2023-03-16T16:45:32.903Z", "content": "Add event_id column to your submission file df[\"event_id\"] = df.index to current submission file. See if it works
", "rawMarkdown": "Add event_id column to your submission file df[\"event_id\"] = df.index to current submission file. See if it works", "replies": [ { "id": 2184982, "postDate": "2023-03-16T18:11:07.993Z", "content": "Thanks for your suggestion.
\n![\"new_submission\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F11662325%2F5d5a421bc0269a48de2c1d5598473bcf%2Foutput.jpg?generation=1678990227173277&alt=media\")
Unfortunately, it didn't work. Any other ideas?
", "rawMarkdown": "Thanks for your suggestion. \n\n\nUnfortunately, it didn't work. Any other ideas?", "replies": [ { "id": 2185271, "postDate": "2023-03-16T23:06:24.863Z", "content": "Check any notebook that has a score, and you will see what other people did in order to submit successfully.
", "rawMarkdown": "Check any notebook that has a score, and you will see what other people did in order to submit successfully.", "votes": 1 }, { "id": 2185453, "postDate": "2023-03-17T04:37:12.993Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2185703, "postDate": "2023-03-17T08:36:56.993Z", "content": "Hello, thanks for your help. I did, but couldn't figure what was different. I am not running the test batches, that is the issue.
", "rawMarkdown": "Hello, thanks for your help. I did, but couldn't figure what was different. I am not running the test batches, that is the issue." } ] } ] }, { "id": 2184828, "postDate": "2023-03-16T16:19:00.043Z", "content": "Hello all.
\nFirst, thanks for the team hosting this challenge, it's quite interesting.
\nI'd like to ask others participating in this challenge if you can check out if there's anything wrong about my submissions, as I keep having a Submission Scoring Error, and I can't see what's wrong with the output (disregarding the accuracy of results).
\n
\nHere is my kernel: https://www.kaggle.com/code/sofiamatias/icecube-predictions
Thanks a lot in advance and good luck to all!
\nSofia
", "rawMarkdown": "Hello all.\nFirst, thanks for the team hosting this challenge, it's quite interesting.\nI'd like to ask others participating in this challenge if you can check out if there's anything wrong about my submissions, as I keep having a Submission Scoring Error, and I can't see what's wrong with the output (disregarding the accuracy of results).\n\nHere is my kernel: https://www.kaggle.com/code/sofiamatias/icecube-predictions\n\nThanks a lot in advance and good luck to all!\n\nSofia" } ], "comments": [ { "id": 2188949, "author_name": "Taq Seorangpun", "author_url": "", "post_date": "2023-03-20T05:44:59.390000", "content": "Try the following
\nsubmission_df.to_csv('/kaggle/working/submission.csv', index=False)
\nIt seems the directory is missing in your notebook.
As noted by Alex Z your notebook is not looking at all of the test batches. This notebook using catboost would be similiar to yours and you can see how the unknown number of batches that will be in the hidden test can be processed.
\n(https://www.kaggle.com/code/astrung/pcaembed-gpu-catboost-optuna-approach)
\nThat notebook was based on a shared xgboost and it looks like the same code was usable for both a catboost and an xgb approach. Think this one is the origin code for the submission section.
\nIceCube🧊: Neutrino🎆 baseline XGBoost🚀Regression
", "votes": 0, "replies": [ { "id": 2185705, "author_name": "Sofia Matias", "author_url": "", "post_date": "2023-03-17T08:37:15.220000", "content": "Thanks, will look into it.
", "votes": 0, "replies": [] } ] }, { "id": 2185052, "author_name": "Alex Z", "author_url": "", "post_date": "2023-03-16T19:16:13.837000", "content": "Batch 661 is a dummy sample with only three events, the actual test dataset is hidden. When doing inference you should read all the test batches from test/batch_[n].parquet
", "votes": 0, "replies": [ { "id": 2185087, "author_name": "Sofia Matias", "author_url": "", "post_date": "2023-03-16T19:45:12.780000", "content": "Aha, that's it! I'll look into it, thanks!!
", "votes": 0, "replies": [] } ] }, { "id": 2184867, "author_name": "Saalim Zafar", "author_url": "", "post_date": "2023-03-16T16:45:32.903000", "content": "Add event_id column to your submission file df[\"event_id\"] = df.index to current submission file. See if it works
", "votes": 0, "replies": [ { "id": 2184982, "author_name": "Sofia Matias", "author_url": "", "post_date": "2023-03-16T18:11:07.993000", "content": "Thanks for your suggestion.
\n
Unfortunately, it didn't work. Any other ideas?
", "votes": 0, "replies": [ { "id": 2185271, "author_name": "greySnow", "author_url": "", "post_date": "2023-03-16T23:06:24.863000", "content": "Check any notebook that has a score, and you will see what other people did in order to submit successfully.
", "votes": 1, "replies": [] }, { "id": 2185453, "author_name": "", "author_url": "", "post_date": "2023-03-17T04:37:12.993000", "content": "", "votes": 0, "replies": [] }, { "id": 2185703, "author_name": "Sofia Matias", "author_url": "", "post_date": "2023-03-17T08:36:56.993000", "content": "Hello, thanks for your help. I did, but couldn't figure what was different. I am not running the test batches, that is the issue.
", "votes": 0, "replies": [] } ] } ] } ], "raw_markdown_by_id": { "2188949": "Try the following\n`submission_df.to_csv('/kaggle/working/submission.csv', index=False)`\nIt seems the directory is missing in your notebook.\n", "2185532": "As noted by [Alex Z ](https://www.kaggle.com/alexz0)your notebook is not looking at all of the test batches. This notebook using catboost would be similiar to yours and you can see how the unknown number of batches that will be in the hidden test can be processed.\n\n(https://www.kaggle.com/code/astrung/pcaembed-gpu-catboost-optuna-approach)\n\nThat notebook was based on a shared xgboost and it looks like the same code was usable for both a catboost and an xgb approach. Think this one is the origin code for the submission section.\n\n[IceCube🧊: Neutrino🎆 baseline XGBoost🚀Regression](uhttps://www.kaggle.com/code/jirkaborovec/icecube-neutrino-baseline-xgboost-regressionrl)\n", "2185052": "Batch 661 is a dummy sample with only three events, the actual test dataset is hidden. When doing inference you should read all the test batches from test/batch_[n].parquet", "2184867": "Add event_id column to your submission file df[\"event_id\"] = df.index to current submission file. See if it works", "2184828": "Hello all.\nFirst, thanks for the team hosting this challenge, it's quite interesting.\nI'd like to ask others participating in this challenge if you can check out if there's anything wrong about my submissions, as I keep having a Submission Scoring Error, and I can't see what's wrong with the output (disregarding the accuracy of results).\n\nHere is my kernel: https://www.kaggle.com/code/sofiamatias/icecube-predictions\n\nThanks a lot in advance and good luck to all!\n\nSofia" } }