![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3414137%2Fd50ed761ef92da538cba98621fa427ad%2F__results___17_0.png?generation=1576467761110141&alt=media\")
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F3414137%2F855b9fc0f2753923b3be3fb3df58bf92%2F__results___33_1.png?generation=1576470221847012&alt=media\")
I thought about whether to post my solution for a long time because it may not be better than other complicated solutions.
\n\nBut it is really simple and performs well. It’s more like a baseline model. I hope you can learn something from this. : )
\n\nWhat I did:\n- Dropping bin_0\n- Ordinal Encoding ord features\n- One Hot Encoding other features\n- Using logistic regression with ‘liblinear’ solver\n- Tuning C by using optuna
\n\nI did these things except optuna at the beginning of this game.\nI tuned C the last day and trained model all at once instead of using Kfolds. It improved my score to 0.80850
\n\nI’m new to machine learning and this is my first competition. \nThanks to Kaggle and everyone participates in this competition. I learned a lot from you guys, which is more important than the ranking. Thanks for your share!
", "comments": [ { "id": 694350, "author_name": "Robert Stockton", "author_url": "", "post_date": "2019-12-13T13:03:31.100000", "content": "Thanks for the share. Simplicity is good, and you've definitely made a nice start to your Kaggle journey.
\n\nI'll definitely look up optuna, and kick myself for not catching that bin_0 was a confuser rather than an asset.
", "votes": 3, "replies": [] }, { "id": 693951, "author_name": "Advait", "author_url": "", "post_date": "2019-12-13T01:36:37.333000", "content": "Congratulations! What C value did you end up with?
", "votes": 3, "replies": [ { "id": 695012, "author_name": "kanopri", "author_url": "", "post_date": "2019-12-14T12:54:12.263000", "content": "It's 0.12385012421930243, quite close to 0.123456789.\nif I set class_weight={0:1, 1:1.3267279323409777}, which is also optimized by optuna, though I think tuning class_weight is meaningless, C=0.09968474250024324.
", "votes": 2, "replies": [] } ] }, { "id": 694360, "author_name": "Miyabon", "author_url": "", "post_date": "2019-12-13T13:20:12.883000", "content": "Congrats and thanks for sharing! Simple is best!!😄 👍
", "votes": 4, "replies": [] }, { "id": 737294, "author_name": "no1", "author_url": "", "post_date": "2020-02-05T06:35:43.960000", "content": "I'm not able to find the solution link, can anyone please share the link.
", "votes": 1, "replies": [] }, { "id": 698100, "author_name": "aldparis", "author_url": "", "post_date": "2019-12-18T20:00:38.680000", "content": "Congratulations ! And thank you for sharing this simple solution.
", "votes": 1, "replies": [] }, { "id": 696305, "author_name": "", "author_url": "", "post_date": "2019-12-16T12:38:52.470000", "content": "Congrats!
", "votes": 1, "replies": [] }, { "id": 696191, "author_name": "ANH TRAN VT", "author_url": "", "post_date": "2019-12-16T08:36:16.213000", "content": "tks
", "votes": 1, "replies": [] }, { "id": 696038, "author_name": "pandaxxk", "author_url": "", "post_date": "2019-12-16T03:26:34.643000", "content": "Thanks for sharing. And may i ask why dropping bin_0 ?
", "votes": 1, "replies": [ { "id": 696060, "author_name": "kanopri", "author_url": "", "post_date": "2019-12-16T04:25:22.553000", "content": "images from this kernel \nand this kernel\nseems can't learn anything from bin_0\n\n
thanks~
", "votes": 0, "replies": [] } ] }, { "id": 695460, "author_name": "ravi tanwar", "author_url": "", "post_date": "2019-12-15T07:16:53.063000", "content": "congrats !
", "votes": 2, "replies": [] }, { "id": 694772, "author_name": "DiegoJohnson", "author_url": "", "post_date": "2019-12-14T03:52:59.933000", "content": "wonderful solution, congratuation! @kanopri
", "votes": 2, "replies": [] }, { "id": 694520, "author_name": "SelvaMurugan", "author_url": "", "post_date": "2019-12-13T18:28:55.957000", "content": "Congrats!!
", "votes": 2, "replies": [] }, { "id": 693859, "author_name": "", "author_url": "", "post_date": "2019-12-12T21:34:35.150000", "content": "", "votes": 2, "replies": [] }, { "id": 694426, "author_name": "AlbertoDS", "author_url": "", "post_date": "2019-12-13T15:20:10.433000", "content": "Thank you for sharing
", "votes": 4, "replies": [] }, { "id": 694065, "author_name": "GSD", "author_url": "", "post_date": "2019-12-13T05:57:38.940000", "content": "Very simple..Thanks a lot for the share
", "votes": 3, "replies": [] }, { "id": 694745, "author_name": "ma7555", "author_url": "", "post_date": "2019-12-14T02:55:48.253000", "content": "Thanks for sharing and best of luck
", "votes": 2, "replies": [] }, { "id": 1047207, "author_name": "sumairuno", "author_url": "", "post_date": "2020-10-12T11:13:46.417000", "content": "Thank you for sharing😊
", "votes": 0, "replies": [] } ], "comment_count": 19, "messages": [ { "id": 693663, "postDate": "2019-12-12T16:45:19.737Z", "content": "I thought about whether to post my solution for a long time because it may not be better than other complicated solutions.
\n\nBut it is really simple and performs well. It’s more like a baseline model. I hope you can learn something from this. : )
\n\nWhat I did:\n- Dropping bin_0\n- Ordinal Encoding ord features\n- One Hot Encoding other features\n- Using logistic regression with ‘liblinear’ solver\n- Tuning C by using optuna
\n\nI did these things except optuna at the beginning of this game.\nI tuned C the last day and trained model all at once instead of using Kfolds. It improved my score to 0.80850
\n\nI’m new to machine learning and this is my first competition. \nThanks to Kaggle and everyone participates in this competition. I learned a lot from you guys, which is more important than the ranking. Thanks for your share!
", "rawMarkdown": "I thought about whether to post my solution for a long time because it may not be better than other complicated solutions. \n\n\nBut it is really simple and performs well. It’s more like a baseline model. I hope you can learn something from this. : )\n\nWhat I did:\n- Dropping bin_0\n- Ordinal Encoding ord features\n- One Hot Encoding other features\n- Using logistic regression with ‘liblinear’ solver\n- Tuning C by using [optuna](https://www.kaggle.com/cuijamm/simple-onehot-logisticregression-score-0-80801)\n\nI did these things except optuna at the beginning of this game.\nI tuned C the last day and trained model all at once instead of using Kfolds. It improved my score to 0.80850\n\nI’m new to machine learning and this is my first competition. \nThanks to Kaggle and everyone participates in this competition. I learned a lot from you guys, which is more important than the ranking. Thanks for your share!\n", "votes": 53 }, { "id": 694350, "postDate": "2019-12-13T13:03:31.100Z", "content": "Thanks for the share. Simplicity is good, and you've definitely made a nice start to your Kaggle journey.
\n\nI'll definitely look up optuna, and kick myself for not catching that bin_0 was a confuser rather than an asset.
", "rawMarkdown": "Thanks for the share. Simplicity is good, and you've definitely made a nice start to your Kaggle journey.\n\nI'll definitely look up optuna, and kick myself for not catching that bin_0 was a confuser rather than an asset.\n", "votes": 3 }, { "id": 693951, "postDate": "2019-12-13T01:36:37.333Z", "content": "Congratulations! What C value did you end up with?
", "rawMarkdown": "Congratulations! What C value did you end up with?", "votes": 3, "replies": [ { "id": 695012, "postDate": "2019-12-14T12:54:12.263Z", "content": "It's 0.12385012421930243, quite close to 0.123456789.\nif I set class_weight={0:1, 1:1.3267279323409777}, which is also optimized by optuna, though I think tuning class_weight is meaningless, C=0.09968474250024324.
", "rawMarkdown": "It's 0.12385012421930243, quite close to 0.123456789.\nif I set class_weight={0:1, 1:1.3267279323409777}, which is also optimized by optuna, though I think tuning class_weight is meaningless, C=0.09968474250024324.", "votes": 2 } ] }, { "id": 694360, "postDate": "2019-12-13T13:20:12.883Z", "content": "Congrats and thanks for sharing! Simple is best!!😄 👍
", "rawMarkdown": "Congrats and thanks for sharing! Simple is best!!😄 👍 ", "votes": 4 }, { "id": 737294, "postDate": "2020-02-05T06:35:43.960Z", "content": "I'm not able to find the solution link, can anyone please share the link.
", "rawMarkdown": "I'm not able to find the solution link, can anyone please share the link.", "votes": 1 }, { "id": 698100, "postDate": "2019-12-18T20:00:38.680Z", "content": "Congratulations ! And thank you for sharing this simple solution.
", "rawMarkdown": "Congratulations ! And thank you for sharing this simple solution.", "votes": 1 }, { "id": 696305, "postDate": "2019-12-16T12:38:52.470Z", "content": "Congrats!
", "rawMarkdown": "Congrats!", "votes": 1 }, { "id": 696191, "postDate": "2019-12-16T08:36:16.213Z", "content": "tks
", "rawMarkdown": "tks ", "votes": 1 }, { "id": 696038, "postDate": "2019-12-16T03:26:34.643Z", "content": "Thanks for sharing. And may i ask why dropping bin_0 ?
", "rawMarkdown": "Thanks for sharing. And may i ask why dropping bin_0 ?", "votes": 1, "replies": [ { "id": 696060, "postDate": "2019-12-16T04:25:22.553Z", "content": "images from this kernel \nand this kernel\nseems can't learn anything from bin_0\n\n
thanks~
", "rawMarkdown": "thanks~" } ] }, { "id": 695460, "postDate": "2019-12-15T07:16:53.063Z", "content": "congrats !
", "rawMarkdown": "congrats !", "votes": 2 }, { "id": 694772, "postDate": "2019-12-14T03:52:59.933Z", "content": "wonderful solution, congratuation! @kanopri
", "rawMarkdown": "wonderful solution, congratuation! @kanopri ", "votes": 2 }, { "id": 694520, "postDate": "2019-12-13T18:28:55.957Z", "content": "Congrats!!
", "rawMarkdown": "Congrats!!", "votes": 2 }, { "id": 693859, "postDate": "2019-12-12T21:34:35.150Z", "rawMarkdown": "", "votes": 2, "isDeleted": true }, { "id": 694426, "postDate": "2019-12-13T15:20:10.433Z", "content": "Thank you for sharing
", "rawMarkdown": "Thank you for sharing", "votes": 4 }, { "id": 694065, "postDate": "2019-12-13T05:57:38.940Z", "content": "Very simple..Thanks a lot for the share
", "rawMarkdown": "Very simple..Thanks a lot for the share", "votes": 3 }, { "id": 694745, "postDate": "2019-12-14T02:55:48.253Z", "content": "Thanks for sharing and best of luck
", "rawMarkdown": "Thanks for sharing and best of luck", "votes": 2 }, { "id": 1047207, "postDate": "2020-10-12T11:13:46.417Z", "content": "Thank you for sharing😊
", "rawMarkdown": "Thank you for sharing😊" } ], "complete": true, "source": "live", "fetched_at": "2026-10-19T17:33:10Z", "competition_id": 14999, "competition_slug": "cat-in-the-dat", "team_id": 3574123, "solution_url": "/competitions/cat-in-the-dat/writeups/hello-world-1st-place-simple-solution" }