Hello, I have a question for the competition sponsors. I am affiliated with a company that would like to participate but we are not sure whether we will be able to share our code. Can we still participate, assuming we 'opt-out' of being eligible for the prize?
", "messages": [ { "id": "2034845", "postDate": "11/18/2022 13:15:39", "content": "Hello, I have a question for the competition sponsors. I am affiliated with a company that would like to participate but we are not sure whether we will be able to share our code. Can we still participate, assuming we 'opt-out' of being eligible for the prize?
", "rawMarkdown": "Hello, I have a question for the competition sponsors. I am affiliated with a company that would like to participate but we are not sure whether we will be able to share our code. Can we still participate, assuming we 'opt-out' of being eligible for the prize?", "votes": null }, { "id": "2038760", "postDate": "11/21/2022 16:12:42", "content": "Hey @ethanmd0519, thanks for your question! Unfortunately, I have to tell you that we are not making any exceptions, and all participants need to open-source their winning solutions under the MIT License. We want everyone in the research community to profit from the dataset and the outcomes of this competition; therefore, closed-source solutions are not an option.
\nIf open-sourcing your solution is impossible, and you still want to evaluate your algorithm against the leaderboard, you can use our offline evaluation scripts from our GitHub repo. @radek1 and @cdeotte have demonstrated that the local validation correlates highly with the leaderboard here on Kaggle and that you can use it as an indicator of your relative performance.
\nNevertheless, I hope you can convince your coworkers to participate in the competition, and we'll see you on the leaderboard soon 😉
", "rawMarkdown": "Hey @ethanmd0519, thanks for your question! Unfortunately, I have to tell you that we are not making any exceptions, and all participants need to open-source their winning solutions under the [MIT License](https://choosealicense.com/licenses/mit/). We want everyone in the research community to profit from the dataset and the outcomes of this competition; therefore, closed-source solutions are not an option.\n\nIf open-sourcing your solution is impossible, and you still want to evaluate your algorithm against the leaderboard, you can use our offline evaluation scripts from our [GitHub](https://github.com/otto-de/recsys-dataset) repo. @radek1 and @cdeotte have [demonstrated](https://www.kaggle.com/competitions/otto-recommender-system/discussion/364991) that the local validation correlates highly with the leaderboard here on Kaggle and that you can use it as an indicator of your relative performance.\n\nNevertheless, I hope you can convince your coworkers to participate in the competition, and we'll see you on the leaderboard soon 😉", "votes": null } ], "comments": [ { "id": 2038760, "author_name": "pnormann", "author_url": "", "post_date": "11/21/2022 16:12:42", "content": "Hey @ethanmd0519, thanks for your question! Unfortunately, I have to tell you that we are not making any exceptions, and all participants need to open-source their winning solutions under the MIT License. We want everyone in the research community to profit from the dataset and the outcomes of this competition; therefore, closed-source solutions are not an option.
\nIf open-sourcing your solution is impossible, and you still want to evaluate your algorithm against the leaderboard, you can use our offline evaluation scripts from our GitHub repo. @radek1 and @cdeotte have demonstrated that the local validation correlates highly with the leaderboard here on Kaggle and that you can use it as an indicator of your relative performance.
\nNevertheless, I hope you can convince your coworkers to participate in the competition, and we'll see you on the leaderboard soon 😉
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2034845": "Hello, I have a question for the competition sponsors. I am affiliated with a company that would like to participate but we are not sure whether we will be able to share our code. Can we still participate, assuming we 'opt-out' of being eligible for the prize?", "2038760": "Hey @ethanmd0519, thanks for your question! Unfortunately, I have to tell you that we are not making any exceptions, and all participants need to open-source their winning solutions under the [MIT License](https://choosealicense.com/licenses/mit/). We want everyone in the research community to profit from the dataset and the outcomes of this competition; therefore, closed-source solutions are not an option.\n\nIf open-sourcing your solution is impossible, and you still want to evaluate your algorithm against the leaderboard, you can use our offline evaluation scripts from our [GitHub](https://github.com/otto-de/recsys-dataset) repo. @radek1 and @cdeotte have [demonstrated](https://www.kaggle.com/competitions/otto-recommender-system/discussion/364991) that the local validation correlates highly with the leaderboard here on Kaggle and that you can use it as an indicator of your relative performance.\n\nNevertheless, I hope you can convince your coworkers to participate in the competition, and we'll see you on the leaderboard soon 😉" }, "source": "meta" }