What is the licensing on this datset (e.g., public domain, CC BY-SA, CC BY-NC, etc.)?
\nCan I use it for commercial use (monetized personal blog posts / data analysis portfolio)
Usually, we can find the dataset license on the dataset's page. Since this is a competition dataset, it doesn't have a dataset page and thus I can't see the dataset license.
", "messages": [ { "id": 1981509, "postDate": "2022-10-10T20:21:53.620Z", "content": "What is the licensing on this datset (e.g., public domain, CC BY-SA, CC BY-NC, etc.)?
\nCan I use it for commercial use (monetized personal blog posts / data analysis portfolio)
Usually, we can find the dataset license on the dataset's page. Since this is a competition dataset, it doesn't have a dataset page and thus I can't see the dataset license.
", "rawMarkdown": "What is the licensing on this datset (e.g., public domain, CC BY-SA, CC BY-NC, etc.)?\nCan I use it for commercial use (monetized personal blog posts / data analysis portfolio)\n\nUsually, we can find the dataset license on the dataset's page. Since this is a competition dataset, it doesn't have a dataset page and thus I can't see the dataset license.", "votes": 13 }, { "id": 1981528, "postDate": "2022-10-10T20:37:45.407Z", "content": "Hi Leonie,
\nThe Competition Data License is outlined in the Competition Rules.
\nWhile we cannot provide legal advice/license interpretation for you, I've pasted the relevant section below for your reference:
\n\n", "rawMarkdown": "Hi Leonie,\n\nThe Competition Data License is outlined in the Competition Rules.\n\nWhile we cannot provide legal advice/license interpretation for you, I've pasted the relevant section below for your reference:\n\n>7. COMPETITION DATA.\n\"Competition Data\" means the data or datasets available from the Competition Website for the purpose of use in the Competition, including any prototype or executable code provided on the Competition Website. The Competition Data will contain private and public test sets. Which data belongs to which set will not be made available to participants.\n\n> A. Data Access and Use. You may access and use the Competition Data (excluding the Race Data) for non-commercial purposes only, including for participating in the Competition and on Kaggle.com forums, and for academic research and education. The Competition Sponsor reserves the right to disqualify any participant who uses the Competition Data other than as permitted by the Competition Website and these Rules. Race Data may only be used only for participating in the Competition.\n\n> B. Data Security. You agree to use reasonable and suitable measures to prevent persons who have not formally agreed to these Rules from gaining access to the Competition Data. You agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Competition Data to any party not participating in the Competition. You agree to notify Kaggle immediately upon learning of any possible unauthorized transmission of or unauthorized access to the Competition Data and agree to work with Kaggle to rectify any unauthorized transmission or access.\n\n> C. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit you other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).\n\n> D. Race Data. You agree that you will only use the Race Data in your participation of the Competition and for no other purpose. Upon the conclusion of the Competition, you agree that your limited license to use the Race Data shall terminate and that you will delete and purge from your possession and control all files, documents, records, databases and the like containing the Race Data or any portions thereof.", "votes": 2, "replies": [ { "id": 1981942, "postDate": "2022-10-11T06:31:03.457Z", "content": "\n
\n- COMPETITION DATA.
\n
\n \"Competition Data\" means the data or datasets available from the Competition Website for the purpose of use in the Competition, including any prototype or executable code provided on the Competition Website. The Competition Data will contain private and public test sets. Which data belongs to which set will not be made available to participants.A. Data Access and Use. You may access and use the Competition Data (excluding the Race Data) for non-commercial purposes only, including for participating in the Competition and on Kaggle.com forums, and for academic research and education. The Competition Sponsor reserves the right to disqualify any participant who uses the Competition Data other than as permitted by the Competition Website and these Rules. Race Data may only be used only for participating in the Competition.
\nB. Data Security. You agree to use reasonable and suitable measures to prevent persons who have not formally agreed to these Rules from gaining access to the Competition Data. You agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Competition Data to any party not participating in the Competition. You agree to notify Kaggle immediately upon learning of any possible unauthorized transmission of or unauthorized access to the Competition Data and agree to work with Kaggle to rectify any unauthorized transmission or access.
\nC. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit you other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).
\nD. Race Data. You agree that you will only use the Race Data in your participation of the Competition and for no other purpose. Upon the conclusion of the Competition, you agree that your limited license to use the Race Data shall terminate and that you will delete and purge from your possession and control all files, documents, records, databases and the like containing the Race Data or any portions thereof.
\n
Thanks for the clarification. I will keep it in mind to check the competition rules next time.
", "rawMarkdown": "Thanks for the clarification. I will keep it in mind to check the competition rules next time." }, { "id": 2018498, "postDate": "2022-11-05T18:32:50.937Z", "content": "Hi NYRA Team, @addisonhoward
\nHope you are well. I require clarification on Point C (External Data) in the rules described for the competition.
\nCurrently, to train my model, I have utilized a publicly available dataset. As per the rules, I have to \"ensure the External Data is publicly available to use by all participants of the Competition for purposes of the competition at no cost to the other participants\".
\nDoes this mean I have to explicitly share or publish the dataset to my account or just confirm this external dataset that I am using is available for free? Please let me know. Thank you.
\nTo note, I have got the dataset at no cost, and is available to use for all at this link here
", "rawMarkdown": "Hi NYRA Team, @addisonhoward \n\nHope you are well. I require clarification on **Point C** (**External Data**) in the rules described for the competition.\n\nCurrently, to train my model, I have utilized a publicly available dataset. As per the rules, I have to \"**ensure the External Data is publicly available to use by all participants of the Competition for purposes of the competition at no cost to the other participants**\".\n\n**Does this mean I have to explicitly share or publish the dataset to my account** or **just confirm this external dataset that I am using is available for free?** Please let me know. Thank you.\n\nTo note, I have got the dataset at no cost, and is available to use for all at this link [here](https://www.equibase.com/)", "votes": 1 }, { "id": 2018538, "postDate": "2022-11-05T19:30:24.253Z", "content": "Hi Kush,
\nFor some context, this rule is intended to make a machine learning competition really about machine learning. Many of us know that many models can become more accurate simply through having better data. Therefore, we want the competition to as much about the best \"data science\" as possible, and not as much about \"who has the best proprietary data.\" If the data is available at no cost for anyone - it's free for you to use and you don't need to post/share to your account (unless you want to, to help out other Kagglers!)
\nHope that helps!
", "rawMarkdown": "Hi Kush,\n\nFor some context, this rule is intended to make a machine learning competition really about machine learning. Many of us know that many models can become more accurate simply through having better data. Therefore, we want the competition to as much about the best \"data science\" as possible, and not as much about \"who has the best proprietary data.\" If the data is available at no cost for anyone - it's free for you to use and you don't need to post/share to your account (unless you want to, to help out other Kagglers!)\n\nHope that helps!", "votes": 1 }, { "id": 2020134, "postDate": "2022-11-07T07:49:28.520Z", "content": "Thank you so much for the information I appreciate it. @addisonhoward
", "rawMarkdown": "Thank you so much for the information I appreciate it. @addisonhoward " }, { "id": 2028863, "postDate": "2022-11-14T09:13:10.883Z", "content": "@addisonhoward does this also hold true for commercial software? I am a colleague of the 4 kagglers who participated as a team, and I was told about the competition and their work just recently, so I registered on Kaggle to read their contribution. I came stumbled on a notebook that makes use of \"black box\" software, is that what a competition is about? It's unclear how he came up with the findings he mentions, he may have already entered anything into this software, and nobody will be able to check it unless they purchase the software! I'm not going to judge the work (because then there would be a lot to say…), but I'm talking about the fact that this notebook is the only one (I've already checked all the notebooks in the code section) that throws a file \"whose provenance is unknown\" and claims to have used H2O.ai software. Even when I tried to use it by getting in touch with H2O.ai, I was informed that prices vary depending on the features, the training, etc. Kaggle, then please stay true to your mission and don't promote the business objectives of select firms. Another interesting thing: this notebook, the sameI mentioned, along with three others (one of whom just updated the notebook of his submission today), even kept working on the notebook after the submission deadline had passed, he continued to work on it while the deadline passed 19 hours ago. Otherwise, everyone would be permitted to act similarly, which is unacceptable! However, why set deadlines? for people to act abusively and disrespect everything : deadline, other competitors,etc. And also, I am a colleague of several others who have already participated in NFL competitions for four consecutive years, and this post_NFL_bias_concern is able to say everything about all the overruns and the lack of credibility that all the kagglers have started to understand.
", "rawMarkdown": "@addisonhoward does this also hold true for commercial software? I am a colleague of the 4 kagglers who participated as a team, and I was told about the competition and their work just recently, so I registered on Kaggle to read their contribution. I came stumbled on a notebook that makes use of \"black box\" software, is that what a competition is about? It's unclear how he came up with the findings he mentions, he may have already entered anything into this software, and nobody will be able to check it unless they purchase the software! I'm not going to judge the work (because then there would be a lot to say...), but I'm talking about the fact that this [notebook](https://www.kaggle.com/code/jpmiller/horse-racing-welcome-to-the-machine) is the only one (I've already checked all the notebooks in the code section) that throws a file \"whose provenance is unknown\" and claims to have used H2O.ai software. Even when I tried to use it by getting in touch with H2O.ai, I was informed that prices vary depending on the features, the training, etc. Kaggle, then please stay true to your mission and don't promote the business objectives of select firms. Another interesting thing: this [notebook](https://www.kaggle.com/code/jpmiller/horse-racing-welcome-to-the-machine), the sameI mentioned, along with three others (one of whom just updated the notebook of his submission today), even kept working on the notebook after the submission deadline had passed, he continued to work on it while the deadline passed 19 hours ago. Otherwise, everyone would be permitted to act similarly, which is unacceptable! However, why set deadlines? for people to act abusively and disrespect everything : deadline, other competitors,etc. And also, I am a colleague of several others who have already participated in NFL competitions for four consecutive years, and this [post_NFL_bias_concern](https://www.kaggle.com/competitions/nfl-big-data-bowl-2023/discussion/359200) is able to say everything about all the overruns and the lack of credibility that all the kagglers have started to understand. ", "votes": -2 }, { "id": 2029663, "postDate": "2022-11-14T19:26:51.473Z", "content": "@alex261 Thank you so much for explaining everything so clearly. I just read the notebook you're talking about, and what you say is really true. I also just read the NFL post you mentioned (I'm not a huge football fan, becuase I've been aware of the NFL competition for a while, but I never considered entering to learn what you mentioned!), but I'm shocked! I hope Kaggle @addisonhoward and NYTHA/NYRA @nythanyra are aware of this!
", "rawMarkdown": "@alex261 Thank you so much for explaining everything so clearly. I just read the notebook you're talking about, and what you say is really true. I also just read the NFL post you mentioned (I'm not a huge football fan, becuase I've been aware of the NFL competition for a while, but I never considered entering to learn what you mentioned!), but I'm shocked! I hope Kaggle @addisonhoward and NYTHA/NYRA @nythanyra are aware of this!" }, { "id": 2029680, "postDate": "2022-11-14T19:45:16.600Z", "content": "Hi @alex261 , Competition rules 11. Winners Obligations, A) states:
\n\n", "rawMarkdown": "Hi @alex261 , Competition rules 11. Winners Obligations, A) states:\n\n> As a condition to being awarded a Prize, a Prize winner must fulfill the following obligations:\n(a) deliver to the Competition Sponsor the final model's software code as used to generate the winning Submission and associated documentation. The delivered software code should follow these documentation guidelines, must be capable of generating the winning Submission, and contain a description of resources required to build and/or run the executable code successfully. **If the final model’s software code includes generally commercially available software that is not owned by you, but that can be procured by the Competition Sponsor without undue expense, then instead of delivering the code for that software to the Competition Sponsor, you must identify that software, method for procuring it, and any parameters or other information necessary to replicate the winning Submission;**\n\n\n", "votes": 2 }, { "id": 2029682, "postDate": "2022-11-14T19:46:44.643Z", "content": "As a condition to being awarded a Prize, a Prize winner must fulfill the following obligations:
\n
\n (a) deliver to the Competition Sponsor the final model's software code as used to generate the winning Submission and associated documentation. The delivered software code should follow these documentation guidelines, must be capable of generating the winning Submission, and contain a description of resources required to build and/or run the executable code successfully. If the final model’s software code includes generally commercially available software that is not owned by you, but that can be procured by the Competition Sponsor without undue expense, then instead of delivering the code for that software to the Competition Sponsor, you must identify that software, method for procuring it, and any parameters or other information necessary to replicate the winning Submission;
Further, I'll note that we don't prohibit users from editing their notebooks after they're submitted, but the version to be judged is the latest one as of the submission deadline.
", "rawMarkdown": "Further, I'll note that we don't prohibit users from editing their notebooks after they're submitted, but **the version to be judged is the latest one as of the submission deadline**.", "votes": 3 }, { "id": 2029758, "postDate": "2022-11-14T21:43:21.263Z", "content": "@addisonhoward What you described only applies to leaderboard competitions because, when someone wins, there are no judges; it is his score that determines it (it's automatic; if his score exceeds the others and he is placed among the top 3 (depending on the competition), he is a winner, even before the host knows), and also because it cannot be known in advance which method, which model, or which software was used (and, in general, the winners never disclose their code before or after, only a solution writeup). As a result, to win the prize, they are required to provide the code, the algorithm, potential software, and other information to the competition host. However, in these analytics competitions, everyone publishes their notebook first, and to the best of my understanding, in analytics competitions, the judges are unable to get in touch with the potential winners before announcing the winners. Therefore, I don't believe the judges will get in touch with him (for example) to have him evaluate the software and first see if he replicates the outcomes.
\nHis last NFL-winning notebook this one, mouse over +4, and you will see all the names (at least two team members have already been recruited by the NFL, and both are already the hosts of the NFL 2023 version (as well as NFL 2022): Michael Lopez and Thompson Bliss)
", "rawMarkdown": "@addisonhoward What you described only applies to leaderboard competitions because, when someone wins, there are no judges; it is his score that determines it (it's automatic; if his score exceeds the others and he is placed among the top 3 (depending on the competition), he is a winner, even before the host knows), and also because it cannot be known in advance which method, which model, or which software was used (and, in general, the winners never disclose their code before or after, only a solution writeup). As a result, to win the prize, they are required to provide the code, the algorithm, potential software, and other information to the competition host. However, in these analytics competitions, everyone publishes their notebook first, and to the best of my understanding, in analytics competitions, the judges are unable to get in touch with the potential winners before announcing the winners. Therefore, I don't believe the judges will get in touch with him (for example) to have him evaluate the software and first see if he replicates the outcomes.\n\n His last NFL-winning notebook [this one](https://www.kaggle.com/code/jpmiller/augmented-reality-for-kickoffs-and-punts/notebook), mouse over +4, and you will see all the names (at least two team members have already been recruited by the NFL, and both are already the hosts of the NFL 2023 version (as well as NFL 2022): Michael Lopez and Thompson Bliss)", "votes": -3 }, { "id": 2029759, "postDate": "2022-11-14T21:50:40.343Z", "content": "@addisonhoward and for the 2nd part, of the answer, I say It's unreasonable to state that \"the version to be judged is the latest one as of the submission deadline.\" If someone has done five or six updates (or even one), in your opinion, the judges will now look into the earlier versions, that sounds like a little science fiction to me! (but what is the competitor's update for if not to make his code better for the judges? to satisfy the kagglers, for example?)
", "rawMarkdown": "@addisonhoward and for the 2nd part, of the answer, I say It's unreasonable to state that **\"the version to be judged is the latest one as of the submission deadline.\"** If someone has done five or six updates (or even one), in your opinion, the judges will now look into the earlier versions, that sounds like a little science fiction to me! (but what is the competitor's update for if not to make his code better for the judges? to satisfy the kagglers, for example?)", "votes": -2 }, { "id": 2029778, "postDate": "2022-11-14T22:30:00.097Z", "content": "Hi @alex261 -
\nI'll address your concerns here:
\n1) Winners are considered preliminary until the solutions can be validated. Should the code not satisfy the criteria of the rules, the hosts have the right to move down to the next best solution for an award.
\n2) In the case of the NFL-winning notebook having the NFL hosts on it, that is because in that competition, the requirements of winning included adding the host to the winning notebook as a collaborator.
\n3) Even prior to seeing your comment, we notified the host about any concerns around version control on submitted notebooks.
You are welcome to raise questions about the process, but making assumptions about the intent or integrity of other participants is not acceptable.
", "rawMarkdown": "Hi @alex261 - \n\nI'll address your concerns here:\n\n1) Winners are considered preliminary until the solutions can be validated. Should the code not satisfy the criteria of the rules, the hosts have the right to move down to the next best solution for an award.\n2) In the case of the NFL-winning notebook having the NFL hosts on it, that is because in that competition, the requirements of winning included adding the host to the winning notebook as a collaborator. \n3) Even prior to seeing your comment, we notified the host about any concerns around version control on submitted notebooks. \n\nYou are welcome to raise questions about the process, but making assumptions about the intent or integrity of other participants is not acceptable.", "votes": 2 }, { "id": 2029785, "postDate": "2022-11-14T23:04:33.713Z", "content": "@addisonhoward In any case, we'll see. I'm not the only one who has said that; you should observe how heated the discussions are on the forums for every NFL competition when the results are announced (even from the most famous kaggle GrandMasters!). But the issue here is that if everything were transparent and obvious, no one would make assumptions about the motivation or character of others… Additionally, it's your responsibility to analyze your procedures, policies, and everything else to determine why users of the forums frequently engage in such behavior and not only to silence them by saying it is not acceptable. Thanks.
\nI'm not going to discuss this NFL-winning notebook yet, which included the two NFL hosts; otherwise, I'm not going to finish (anyway, I'm not convinced of what has been said).
\nAnd One of the absurdities you need to look over is the fact that you don't allow people to use data they've paid for (but they must publish it to everyone), yet you allow them to use software from a company that charges and isn't available to everyone. Complete nonsense! Thanks.
", "rawMarkdown": "@addisonhoward In any case, we'll see. I'm not the only one who has said that; you should observe how heated the discussions are on the forums for every NFL competition when the results are announced (even from the most famous kaggle GrandMasters!). But the issue here is that if everything were transparent and obvious, no one would make assumptions about the motivation or character of others... Additionally, it's your responsibility to analyze your procedures, policies, and everything else to determine why users of the forums frequently engage in such behavior and not only to silence them by saying it is not acceptable. Thanks.\n\nI'm not going to discuss this NFL-winning notebook yet, which included the two NFL hosts; otherwise, I'm not going to finish (anyway, I'm not convinced of what has been said).\n\nAnd One of the absurdities you need to look over is the fact that you don't allow people to use data they've paid for (but they must publish it to everyone), yet **you allow them to use software from a company that charges and isn't available to everyone**. Complete nonsense! Thanks.", "votes": -4 }, { "id": 2031352, "postDate": "2022-11-16T02:54:14.370Z", "content": "(Edited)
\nWhy would someone who has been around Kaggle - for 4 years - register a few days ago just to read their colleagues' submission? And then spend so much energy knocking down the best submissions of other teams? And not upvote the notebook he came to read?
This is an analytics challenge, not a ML competition - using commercial tools is in the best interest of the host if it meets the rules and provides better analysis. (Full disclosure: we also used Excel😁.) We'll gladly provide the open source end-to-end solution to our hosts, data prep included. As they say on TV, \"No purchase necessary!\"
", "rawMarkdown": "(Edited)\nWhy would someone who has been around Kaggle - for 4 years - register a few days ago just to read their colleagues' submission? And then spend so much energy knocking down the best submissions of other teams? And not upvote the notebook he came to read?\n\nThis is an analytics challenge, not a ML competition - using commercial tools is in the best interest of the host if it meets the rules and provides better analysis. (Full disclosure: we also used Excel😁.) We'll gladly provide the open source end-to-end solution to our hosts, data prep included. As they say on TV, \"No purchase necessary!\"\n", "votes": 1 }, { "id": 2031438, "postDate": "2022-11-16T04:45:32.340Z", "rawMarkdown": "", "isDeleted": true }, { "id": 2031780, "postDate": "2022-11-16T08:39:23.507Z", "content": "@jpmiller If so, you would not have to use paid H2O.ai software
\nThe word file does not add anything , you give the path of the training file on your H2O.ai machine , validation: no provided, test: no provided and IMHO even a beginner student can train a model with inputs and give values! Is that machine learning? Already rmse is very high on training
\nAnd, since you used software that only certain people from H2O.ai can access, you are competing on behalf of H2O.ai; your company is the participant, not you as a data scientist. If that's the goal, why do a competition on kaggle, they should have gone to H2O.ai or any other AI firm. I don't think that kaggle, through the competition, wants to promote business objectives of H2O.ai for example
Hi Leonie,
\nThe Competition Data License is outlined in the Competition Rules.
\nWhile we cannot provide legal advice/license interpretation for you, I've pasted the relevant section below for your reference:
\n\n", "votes": 2, "replies": [ { "id": 1981942, "author_name": "Leonie", "author_url": "", "post_date": "2022-10-11T06:31:03.457000", "content": "\n
\n- COMPETITION DATA.
\n
\n \"Competition Data\" means the data or datasets available from the Competition Website for the purpose of use in the Competition, including any prototype or executable code provided on the Competition Website. The Competition Data will contain private and public test sets. Which data belongs to which set will not be made available to participants.A. Data Access and Use. You may access and use the Competition Data (excluding the Race Data) for non-commercial purposes only, including for participating in the Competition and on Kaggle.com forums, and for academic research and education. The Competition Sponsor reserves the right to disqualify any participant who uses the Competition Data other than as permitted by the Competition Website and these Rules. Race Data may only be used only for participating in the Competition.
\nB. Data Security. You agree to use reasonable and suitable measures to prevent persons who have not formally agreed to these Rules from gaining access to the Competition Data. You agree not to transmit, duplicate, publish, redistribute or otherwise provide or make available the Competition Data to any party not participating in the Competition. You agree to notify Kaggle immediately upon learning of any possible unauthorized transmission of or unauthorized access to the Competition Data and agree to work with Kaggle to rectify any unauthorized transmission or access.
\nC. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit you other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).
\nD. Race Data. You agree that you will only use the Race Data in your participation of the Competition and for no other purpose. Upon the conclusion of the Competition, you agree that your limited license to use the Race Data shall terminate and that you will delete and purge from your possession and control all files, documents, records, databases and the like containing the Race Data or any portions thereof.
\n
Thanks for the clarification. I will keep it in mind to check the competition rules next time.
", "votes": 0, "replies": [] }, { "id": 2018498, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2022-11-05T18:32:50.937000", "content": "Hi NYRA Team, @addisonhoward
\nHope you are well. I require clarification on Point C (External Data) in the rules described for the competition.
\nCurrently, to train my model, I have utilized a publicly available dataset. As per the rules, I have to \"ensure the External Data is publicly available to use by all participants of the Competition for purposes of the competition at no cost to the other participants\".
\nDoes this mean I have to explicitly share or publish the dataset to my account or just confirm this external dataset that I am using is available for free? Please let me know. Thank you.
\nTo note, I have got the dataset at no cost, and is available to use for all at this link here
", "votes": 1, "replies": [] }, { "id": 2018538, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-11-05T19:30:24.253000", "content": "Hi Kush,
\nFor some context, this rule is intended to make a machine learning competition really about machine learning. Many of us know that many models can become more accurate simply through having better data. Therefore, we want the competition to as much about the best \"data science\" as possible, and not as much about \"who has the best proprietary data.\" If the data is available at no cost for anyone - it's free for you to use and you don't need to post/share to your account (unless you want to, to help out other Kagglers!)
\nHope that helps!
", "votes": 1, "replies": [] }, { "id": 2020134, "author_name": "Kush Trivedi", "author_url": "", "post_date": "2022-11-07T07:49:28.520000", "content": "Thank you so much for the information I appreciate it. @addisonhoward
", "votes": 0, "replies": [] }, { "id": 2028863, "author_name": "Alex261", "author_url": "", "post_date": "2022-11-14T09:13:10.883000", "content": "@addisonhoward does this also hold true for commercial software? I am a colleague of the 4 kagglers who participated as a team, and I was told about the competition and their work just recently, so I registered on Kaggle to read their contribution. I came stumbled on a notebook that makes use of \"black box\" software, is that what a competition is about? It's unclear how he came up with the findings he mentions, he may have already entered anything into this software, and nobody will be able to check it unless they purchase the software! I'm not going to judge the work (because then there would be a lot to say…), but I'm talking about the fact that this notebook is the only one (I've already checked all the notebooks in the code section) that throws a file \"whose provenance is unknown\" and claims to have used H2O.ai software. Even when I tried to use it by getting in touch with H2O.ai, I was informed that prices vary depending on the features, the training, etc. Kaggle, then please stay true to your mission and don't promote the business objectives of select firms. Another interesting thing: this notebook, the sameI mentioned, along with three others (one of whom just updated the notebook of his submission today), even kept working on the notebook after the submission deadline had passed, he continued to work on it while the deadline passed 19 hours ago. Otherwise, everyone would be permitted to act similarly, which is unacceptable! However, why set deadlines? for people to act abusively and disrespect everything : deadline, other competitors,etc. And also, I am a colleague of several others who have already participated in NFL competitions for four consecutive years, and this post_NFL_bias_concern is able to say everything about all the overruns and the lack of credibility that all the kagglers have started to understand.
", "votes": -2, "replies": [] }, { "id": 2029663, "author_name": "", "author_url": "", "post_date": "2022-11-14T19:26:51.473000", "content": "@alex261 Thank you so much for explaining everything so clearly. I just read the notebook you're talking about, and what you say is really true. I also just read the NFL post you mentioned (I'm not a huge football fan, becuase I've been aware of the NFL competition for a while, but I never considered entering to learn what you mentioned!), but I'm shocked! I hope Kaggle @addisonhoward and NYTHA/NYRA @nythanyra are aware of this!
", "votes": 0, "replies": [] }, { "id": 2029680, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-11-14T19:45:16.600000", "content": "Hi @alex261 , Competition rules 11. Winners Obligations, A) states:
\n\n", "votes": 2, "replies": [] }, { "id": 2029682, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-11-14T19:46:44.643000", "content": "As a condition to being awarded a Prize, a Prize winner must fulfill the following obligations:
\n
\n (a) deliver to the Competition Sponsor the final model's software code as used to generate the winning Submission and associated documentation. The delivered software code should follow these documentation guidelines, must be capable of generating the winning Submission, and contain a description of resources required to build and/or run the executable code successfully. If the final model’s software code includes generally commercially available software that is not owned by you, but that can be procured by the Competition Sponsor without undue expense, then instead of delivering the code for that software to the Competition Sponsor, you must identify that software, method for procuring it, and any parameters or other information necessary to replicate the winning Submission;
Further, I'll note that we don't prohibit users from editing their notebooks after they're submitted, but the version to be judged is the latest one as of the submission deadline.
", "votes": 3, "replies": [] }, { "id": 2029758, "author_name": "Alex261", "author_url": "", "post_date": "2022-11-14T21:43:21.263000", "content": "@addisonhoward What you described only applies to leaderboard competitions because, when someone wins, there are no judges; it is his score that determines it (it's automatic; if his score exceeds the others and he is placed among the top 3 (depending on the competition), he is a winner, even before the host knows), and also because it cannot be known in advance which method, which model, or which software was used (and, in general, the winners never disclose their code before or after, only a solution writeup). As a result, to win the prize, they are required to provide the code, the algorithm, potential software, and other information to the competition host. However, in these analytics competitions, everyone publishes their notebook first, and to the best of my understanding, in analytics competitions, the judges are unable to get in touch with the potential winners before announcing the winners. Therefore, I don't believe the judges will get in touch with him (for example) to have him evaluate the software and first see if he replicates the outcomes.
\nHis last NFL-winning notebook this one, mouse over +4, and you will see all the names (at least two team members have already been recruited by the NFL, and both are already the hosts of the NFL 2023 version (as well as NFL 2022): Michael Lopez and Thompson Bliss)
", "votes": -3, "replies": [] }, { "id": 2029759, "author_name": "Alex261", "author_url": "", "post_date": "2022-11-14T21:50:40.343000", "content": "@addisonhoward and for the 2nd part, of the answer, I say It's unreasonable to state that \"the version to be judged is the latest one as of the submission deadline.\" If someone has done five or six updates (or even one), in your opinion, the judges will now look into the earlier versions, that sounds like a little science fiction to me! (but what is the competitor's update for if not to make his code better for the judges? to satisfy the kagglers, for example?)
", "votes": -2, "replies": [] }, { "id": 2029778, "author_name": "Addison Howard", "author_url": "", "post_date": "2022-11-14T22:30:00.097000", "content": "Hi @alex261 -
\nI'll address your concerns here:
\n1) Winners are considered preliminary until the solutions can be validated. Should the code not satisfy the criteria of the rules, the hosts have the right to move down to the next best solution for an award.
\n2) In the case of the NFL-winning notebook having the NFL hosts on it, that is because in that competition, the requirements of winning included adding the host to the winning notebook as a collaborator.
\n3) Even prior to seeing your comment, we notified the host about any concerns around version control on submitted notebooks.
You are welcome to raise questions about the process, but making assumptions about the intent or integrity of other participants is not acceptable.
", "votes": 2, "replies": [] }, { "id": 2029785, "author_name": "Alex261", "author_url": "", "post_date": "2022-11-14T23:04:33.713000", "content": "@addisonhoward In any case, we'll see. I'm not the only one who has said that; you should observe how heated the discussions are on the forums for every NFL competition when the results are announced (even from the most famous kaggle GrandMasters!). But the issue here is that if everything were transparent and obvious, no one would make assumptions about the motivation or character of others… Additionally, it's your responsibility to analyze your procedures, policies, and everything else to determine why users of the forums frequently engage in such behavior and not only to silence them by saying it is not acceptable. Thanks.
\nI'm not going to discuss this NFL-winning notebook yet, which included the two NFL hosts; otherwise, I'm not going to finish (anyway, I'm not convinced of what has been said).
\nAnd One of the absurdities you need to look over is the fact that you don't allow people to use data they've paid for (but they must publish it to everyone), yet you allow them to use software from a company that charges and isn't available to everyone. Complete nonsense! Thanks.
", "votes": -4, "replies": [] }, { "id": 2031352, "author_name": "JohnM", "author_url": "", "post_date": "2022-11-16T02:54:14.370000", "content": "(Edited)
\nWhy would someone who has been around Kaggle - for 4 years - register a few days ago just to read their colleagues' submission? And then spend so much energy knocking down the best submissions of other teams? And not upvote the notebook he came to read?
This is an analytics challenge, not a ML competition - using commercial tools is in the best interest of the host if it meets the rules and provides better analysis. (Full disclosure: we also used Excel😁.) We'll gladly provide the open source end-to-end solution to our hosts, data prep included. As they say on TV, \"No purchase necessary!\"
", "votes": 1, "replies": [] }, { "id": 2031438, "author_name": "", "author_url": "", "post_date": "2022-11-16T04:45:32.340000", "content": "", "votes": 0, "replies": [] }, { "id": 2031780, "author_name": "Alex261", "author_url": "", "post_date": "2022-11-16T08:39:23.507000", "content": "@jpmiller If so, you would not have to use paid H2O.ai software
\nThe word file does not add anything , you give the path of the training file on your H2O.ai machine , validation: no provided, test: no provided and IMHO even a beginner student can train a model with inputs and give values! Is that machine learning? Already rmse is very high on training
\nAnd, since you used software that only certain people from H2O.ai can access, you are competing on behalf of H2O.ai; your company is the participant, not you as a data scientist. If that's the goal, why do a competition on kaggle, they should have gone to H2O.ai or any other AI firm. I don't think that kaggle, through the competition, wants to promote business objectives of H2O.ai for example