Welcome to the 5th Facebook recruiting competition! This time, we prepared a ranking problem for you, hope you'll enjoy it.
\n\nThis competition is particularly exciting for me personally - I got hired after participating in Facebook's second recruiting challenge. Now, three years later, I am helping to organize a similar challenge.
\n\nWe will maintain the forum presence to answer any data- and recruiting-related questions.
\n\nHave fun and thank you for participating!
", "messages": [ { "id": "119605", "postDate": "05/11/2016 19:45:53", "content": "Welcome to the 5th Facebook recruiting competition! This time, we prepared a ranking problem for you, hope you'll enjoy it.
\n\nThis competition is particularly exciting for me personally - I got hired after participating in Facebook's second recruiting challenge. Now, three years later, I am helping to organize a similar challenge.
\n\nWe will maintain the forum presence to answer any data- and recruiting-related questions.
\n\nHave fun and thank you for participating!
", "rawMarkdown": "Welcome to the 5th Facebook recruiting competition! This time, we prepared a ranking problem for you, hope you'll enjoy it.\r\n\r\nThis competition is particularly exciting for me personally - I got hired after participating in [Facebook's second recruiting challenge][1]. Now, three years later, I am helping to organize a similar challenge.\r\n\r\nWe will maintain the forum presence to answer any data- and recruiting-related questions.\r\n\r\nHave fun and thank you for participating!\r\n\r\n\r\n [1]: https://www.kaggle.com/c/facebook-ii", "votes": null }, { "id": "119608", "postDate": "05/11/2016 20:30:40", "content": "What the "opportunity at interviewing" means? Does it mean that you can just get interview and not the job?
", "rawMarkdown": "What the \"opportunity at interviewing\" means? Does it mean that you can just get interview and not the job?", "votes": null }, { "id": "119610", "postDate": "05/11/2016 20:45:21", "content": "Hi Competitors! Another Facebooker here to answer any questions.
\n\n@Seraphimus - Correct. Unfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion.
", "rawMarkdown": "Hi Competitors! Another Facebooker here to answer any questions.\r\n\r\n@Seraphimus - Correct. Unfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion.", "votes": null }, { "id": "119613", "postDate": "05/11/2016 21:02:52", "content": "I understand legal reasons. But if legal conditions are met, does the winner need to pass through complete Facebook recruitment process or this competition plays the role of the recruitment process?
", "rawMarkdown": "I understand legal reasons. But if legal conditions are met, does the winner need to pass through complete Facebook recruitment process or this competition plays the role of the recruitment process?", "votes": null }, { "id": "119614", "postDate": "05/11/2016 21:11:44", "content": "I don't mean to be intentionally vague - all I can say is that it would indeed replace part of the recruiting process but not all of it. We hope that you can understand that we'd like to get the know the scientists and engineers outside the confines of Kaggle before asking them to join the team. We'd hope that you'd also like to meet us before deciding to work with us. :)
", "rawMarkdown": "I don't mean to be intentionally vague - all I can say is that it would indeed replace part of the recruiting process but not all of it. We hope that you can understand that we'd like to get the know the scientists and engineers outside the confines of Kaggle before asking them to join the team. We'd hope that you'd also like to meet us before deciding to work with us. :)", "votes": null }, { "id": "119615", "postDate": "05/11/2016 21:14:54", "content": "Seraphimus,
\n\nI think the description says it pretty clearly.
\n\n\n\n\nPROCESS
\n \nHighly ranked contestants who indicate their interest will be\n considered by Facebook for interviews, based on their work in the\n competition and additional background.
\n
To sum up, winning this competition is not a one-way ticket to a position at Facebook.
", "rawMarkdown": "Seraphimus,\r\n\r\nI think the description says it pretty clearly.\r\n\r\n> **PROCESS**\r\n> \r\n> Highly ranked contestants who indicate their interest will be\r\n> considered by Facebook for interviews, based on their work in the\r\n> competition and additional background.\r\n\r\nTo sum up, winning this competition is not a one-way ticket to a position at Facebook.", "votes": null }, { "id": "119616", "postDate": "05/11/2016 21:19:29", "content": "There is no chance that I will ever be near the top of the leaderbord because I am a complete newbie. I was just curious. Thank you for your answers!
", "rawMarkdown": "There is no chance that I will ever be near the top of the leaderbord because I am a complete newbie. I was just curious. Thank you for your answers!", "votes": null }, { "id": "119788", "postDate": "05/12/2016 19:44:38", "content": "[quote=Ampersand;119610]\nUnfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion. [/quote]\nDo you provide work visa support for winners or you don't?
", "rawMarkdown": "[quote=Ampersand;119610]\r\nUnfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion. [/quote]\r\nDo you provide work visa support for winners or you don't?", "votes": null }, { "id": "119878", "postDate": "05/13/2016 13:31:53", "content": "started another thread(https://www.kaggle.com/c/facebook-v-predicting-check-ins/forums/t/20923/candidates-must-have-or-be-eligible-to-obtain-authorization-to-work-in) before discovering this one, but i think my question(see below) is also relevant to this thread
\n\nH1B quota ended in april and there would not be any new visas till the april of the next year and it is the only way for most of us to obtain authorization to work in the US, now if someone who needs H1B to legally work in the US scores hight in this competition(and meet all job requirements) does that means that facebook is agree to wait till the next year for candidate to join company or that means that he is not eligible for this competition?
", "rawMarkdown": "started another thread(https://www.kaggle.com/c/facebook-v-predicting-check-ins/forums/t/20923/candidates-must-have-or-be-eligible-to-obtain-authorization-to-work-in) before discovering this one, but i think my question(see below) is also relevant to this thread\r\n\r\nH1B quota ended in april and there would not be any new visas till the april of the next year and it is the only way for most of us to obtain authorization to work in the US, now if someone who needs H1B to legally work in the US scores hight in this competition(and meet all job requirements) does that means that facebook is agree to wait till the next year for candidate to join company or that means that he is not eligible for this competition?", "votes": null }, { "id": "120160", "postDate": "05/15/2016 22:22:59", "content": "awesome dataset build..
\n\nkick ass problem..
\n\nreally hope I can make some time for this
", "rawMarkdown": "awesome dataset build.. \r\n\r\n kick ass problem.. \r\n\r\nreally hope I can make some time for this", "votes": null }, { "id": "120300", "postDate": "05/17/2016 07:34:14", "content": "Has the 50% split of the test data been done randomly?
", "rawMarkdown": "Has the 50% split of the test data been done randomly?", "votes": null }, { "id": "120303", "postDate": "05/17/2016 08:14:49", "content": "[quote=Picarus;120300]
\n\nHas the 50% split of the test data been done randomly?
\n\n[/quote]
\n\nShould be. From the Data page:
\n\nThe train and test dataset are split based on time, and the public/private leaderboard in the test data are split randomly.
", "rawMarkdown": "[quote=Picarus;120300]\r\n\r\nHas the 50% split of the test data been done randomly?\r\n\r\n[/quote]\r\n\r\nShould be. From the Data page:\r\n\r\n*The train and test dataset are split based on time, and the public/private leaderboard in the test data are split randomly.*", "votes": null }, { "id": "120306", "postDate": "05/17/2016 09:34:50", "content": "Did you have a particular place (e.g. downtime SFO) or a generic (kind of) place (e.g. railway platforms) in mind while designing this data set? Or are those details somehow irrelevant to this competition.
", "rawMarkdown": "Did you have a particular place (e.g. downtime SFO) or a generic (kind of) place (e.g. railway platforms) in mind while designing this data set? Or are those details somehow irrelevant to this competition.", "votes": null }, { "id": "120380", "postDate": "05/17/2016 21:21:41", "content": "Hi there!\nThere's an error in submission description\nhttps://www.kaggle.com/c/facebook-v-predicting-check-ins/details/evaluation\nFor every user check in, you must predict a space-delimited list of the businesses they check into. You may submit up to 3 predictions for each check in. The file should contain a header and have the following format:\nrow_id,predictions
\n\nShould be\nrow_id,place_id
\n\nThanks for an interesting competitions.
", "rawMarkdown": "Hi there!\r\nThere's an error in submission description\r\nhttps://www.kaggle.com/c/facebook-v-predicting-check-ins/details/evaluation\r\nFor every user check in, you must predict a space-delimited list of the businesses they check into. You may submit up to 3 predictions for each check in. The file should contain a header and have the following format:\r\nrow_id,**predictions**\r\n\r\nShould be\r\nrow_id,**place_id**\r\n\r\nThanks for an interesting competitions.", "votes": null }, { "id": "120393", "postDate": "05/17/2016 23:57:53", "content": "@DennisSakva, thanks! fixed!
", "rawMarkdown": "DennisSakva, thanks! fixed!", "votes": null }, { "id": "120394", "postDate": "05/18/2016 00:01:43", "content": "It was referring to the contents, not the header value. But yeah, I agree it was confusing upon first read
", "rawMarkdown": "It was referring to the contents, not the header value. But yeah, I agree it was confusing upon first read", "votes": null }, { "id": "120989", "postDate": "05/22/2016 13:17:00", "content": "Hi,\nI'm just curious about the "Requirements" for the Job:\n... 5 years experienced ...
\n\nActually I'm doing my master thesis - I can't really have that much work experienced. :)
\n\nSo, how important are these requirements?
\n\nEdit: Just saw the same question&answer in another thread :)
", "rawMarkdown": "Hi,\r\nI'm just curious about the \"Requirements\" for the Job:\r\n... 5 years experienced ...\r\n\r\nActually I'm doing my master thesis - I can't really have that much work experienced. :)\r\n\r\nSo, how important are these requirements?\r\n\r\nEdit: Just saw the same question&answer in another thread :)", "votes": null }, { "id": "121499", "postDate": "05/26/2016 22:05:40", "content": "I have a question regarding number of predictions it is stated that: "You may submit up to 3 predictions for each check in". Does it mean that some of the test values have less than 3 real predictions. If yes submitting more than real predictions will impact the result.
", "rawMarkdown": "I have a question regarding number of predictions it is stated that: \"You may submit up to 3 predictions for each check in\". Does it mean that some of the test values have less than 3 real predictions. If yes submitting more than real predictions will impact the result.", "votes": null }, { "id": "121500", "postDate": "05/26/2016 22:09:46", "content": "Each test observation has exactly 1 correct prediction: the place where the checkin effectively took place when the data generating simulation was run. You are allowed up to 3 predictions. If your first prediction is right, you get 1/1, if your 2nd prediction is right, you get 1/2, and if your 3rd prediction is right, your get 1/3. Otherwise, you get 0 for the particular observation. Hence, there is no reason not to make 3 predictions per test observation, in the order of likelihood.
", "rawMarkdown": "Each test observation has exactly 1 correct prediction: the place where the checkin effectively took place when the data generating simulation was run. You are allowed up to 3 predictions. If your first prediction is right, you get 1/1, if your 2nd prediction is right, you get 1/2, and if your 3rd prediction is right, your get 1/3. Otherwise, you get 0 for the particular observation. Hence, there is no reason *not* to make 3 predictions per test observation, in the order of likelihood.", "votes": null }, { "id": "121547", "postDate": "05/27/2016 07:44:34", "content": "@Andre Naef Thanks for your clear explanation.
", "rawMarkdown": "Andre Naef Thanks for your clear explanation.", "votes": null }, { "id": "123427", "postDate": "06/11/2016 20:40:15", "content": "What kind of the rank can get the interview opportunity? Top 1 or Top 10? Thanks!
", "rawMarkdown": "What kind of the rank can get the interview opportunity? Top 1 or Top 10? Thanks!", "votes": null }, { "id": "125244", "postDate": "06/28/2016 03:01:55", "content": "Excuse me, Anyone has difficulty uploading the results file? \nI have tried two times, but failed.\nI don't think a file of 100Mb is too large to be uploaded. Is there any tips about uploading results???
", "rawMarkdown": "Excuse me, Anyone has difficulty uploading the results file? \r\nI have tried two times, but failed.\r\nI don't think a file of 100Mb is too large to be uploaded. Is there any tips about uploading results???", "votes": null } ], "comments": [ { "id": 119608, "author_name": "seraphimus", "author_url": "", "post_date": "05/11/2016 20:30:40", "content": "What the "opportunity at interviewing" means? Does it mean that you can just get interview and not the job?
", "votes": null, "replies": [] }, { "id": 119610, "author_name": "youngr11", "author_url": "", "post_date": "05/11/2016 20:45:21", "content": "Hi Competitors! Another Facebooker here to answer any questions.
\n\n@Seraphimus - Correct. Unfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion.
", "votes": null, "replies": [] }, { "id": 119613, "author_name": "seraphimus", "author_url": "", "post_date": "05/11/2016 21:02:52", "content": "I understand legal reasons. But if legal conditions are met, does the winner need to pass through complete Facebook recruitment process or this competition plays the role of the recruitment process?
", "votes": null, "replies": [] }, { "id": 119614, "author_name": "youngr11", "author_url": "", "post_date": "05/11/2016 21:11:44", "content": "I don't mean to be intentionally vague - all I can say is that it would indeed replace part of the recruiting process but not all of it. We hope that you can understand that we'd like to get the know the scientists and engineers outside the confines of Kaggle before asking them to join the team. We'd hope that you'd also like to meet us before deciding to work with us. :)
", "votes": null, "replies": [] }, { "id": 119615, "author_name": "navneethc", "author_url": "", "post_date": "05/11/2016 21:14:54", "content": "Seraphimus,
\n\nI think the description says it pretty clearly.
\n\n\n\n\nPROCESS
\n \nHighly ranked contestants who indicate their interest will be\n considered by Facebook for interviews, based on their work in the\n competition and additional background.
\n
To sum up, winning this competition is not a one-way ticket to a position at Facebook.
", "votes": null, "replies": [] }, { "id": 119616, "author_name": "seraphimus", "author_url": "", "post_date": "05/11/2016 21:19:29", "content": "There is no chance that I will ever be near the top of the leaderbord because I am a complete newbie. I was just curious. Thank you for your answers!
", "votes": null, "replies": [] }, { "id": 119788, "author_name": "rakhlin", "author_url": "", "post_date": "05/12/2016 19:44:38", "content": "[quote=Ampersand;119610]\nUnfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion. [/quote]\nDo you provide work visa support for winners or you don't?
", "votes": null, "replies": [] }, { "id": 119878, "author_name": "valeriur", "author_url": "", "post_date": "05/13/2016 13:31:53", "content": "started another thread(https://www.kaggle.com/c/facebook-v-predicting-check-ins/forums/t/20923/candidates-must-have-or-be-eligible-to-obtain-authorization-to-work-in) before discovering this one, but i think my question(see below) is also relevant to this thread
\n\nH1B quota ended in april and there would not be any new visas till the april of the next year and it is the only way for most of us to obtain authorization to work in the US, now if someone who needs H1B to legally work in the US scores hight in this competition(and meet all job requirements) does that means that facebook is agree to wait till the next year for candidate to join company or that means that he is not eligible for this competition?
", "votes": null, "replies": [] }, { "id": 120160, "author_name": "scharf", "author_url": "", "post_date": "05/15/2016 22:22:59", "content": "awesome dataset build..
\n\nkick ass problem..
\n\nreally hope I can make some time for this
", "votes": null, "replies": [] }, { "id": 120300, "author_name": "picarus", "author_url": "", "post_date": "05/17/2016 07:34:14", "content": "Has the 50% split of the test data been done randomly?
", "votes": null, "replies": [] }, { "id": 120303, "author_name": "peatle", "author_url": "", "post_date": "05/17/2016 08:14:49", "content": "[quote=Picarus;120300]
\n\nHas the 50% split of the test data been done randomly?
\n\n[/quote]
\n\nShould be. From the Data page:
\n\nThe train and test dataset are split based on time, and the public/private leaderboard in the test data are split randomly.
", "votes": null, "replies": [] }, { "id": 120306, "author_name": "navneethc", "author_url": "", "post_date": "05/17/2016 09:34:50", "content": "Did you have a particular place (e.g. downtime SFO) or a generic (kind of) place (e.g. railway platforms) in mind while designing this data set? Or are those details somehow irrelevant to this competition.
", "votes": null, "replies": [] }, { "id": 120380, "author_name": "sakvaua", "author_url": "", "post_date": "05/17/2016 21:21:41", "content": "Hi there!\nThere's an error in submission description\nhttps://www.kaggle.com/c/facebook-v-predicting-check-ins/details/evaluation\nFor every user check in, you must predict a space-delimited list of the businesses they check into. You may submit up to 3 predictions for each check in. The file should contain a header and have the following format:\nrow_id,predictions
\n\nShould be\nrow_id,place_id
\n\nThanks for an interesting competitions.
", "votes": null, "replies": [] }, { "id": 120393, "author_name": "wendykan", "author_url": "", "post_date": "05/17/2016 23:57:53", "content": "@DennisSakva, thanks! fixed!
", "votes": null, "replies": [] }, { "id": 120394, "author_name": "zeraes", "author_url": "", "post_date": "05/18/2016 00:01:43", "content": "It was referring to the contents, not the header value. But yeah, I agree it was confusing upon first read
", "votes": null, "replies": [] }, { "id": 120989, "author_name": "antsek", "author_url": "", "post_date": "05/22/2016 13:17:00", "content": "Hi,\nI'm just curious about the "Requirements" for the Job:\n... 5 years experienced ...
\n\nActually I'm doing my master thesis - I can't really have that much work experienced. :)
\n\nSo, how important are these requirements?
\n\nEdit: Just saw the same question&answer in another thread :)
", "votes": null, "replies": [] }, { "id": 121499, "author_name": "cptzbik", "author_url": "", "post_date": "05/26/2016 22:05:40", "content": "I have a question regarding number of predictions it is stated that: "You may submit up to 3 predictions for each check in". Does it mean that some of the test values have less than 3 real predictions. If yes submitting more than real predictions will impact the result.
", "votes": null, "replies": [] }, { "id": 121500, "author_name": "andrenaef", "author_url": "", "post_date": "05/26/2016 22:09:46", "content": "Each test observation has exactly 1 correct prediction: the place where the checkin effectively took place when the data generating simulation was run. You are allowed up to 3 predictions. If your first prediction is right, you get 1/1, if your 2nd prediction is right, you get 1/2, and if your 3rd prediction is right, your get 1/3. Otherwise, you get 0 for the particular observation. Hence, there is no reason not to make 3 predictions per test observation, in the order of likelihood.
", "votes": null, "replies": [] }, { "id": 121547, "author_name": "cptzbik", "author_url": "", "post_date": "05/27/2016 07:44:34", "content": "@Andre Naef Thanks for your clear explanation.
", "votes": null, "replies": [] }, { "id": 123427, "author_name": "caixiabi", "author_url": "", "post_date": "06/11/2016 20:40:15", "content": "What kind of the rank can get the interview opportunity? Top 1 or Top 10? Thanks!
", "votes": null, "replies": [] }, { "id": 125244, "author_name": "sdcsyuxue", "author_url": "", "post_date": "06/28/2016 03:01:55", "content": "Excuse me, Anyone has difficulty uploading the results file? \nI have tried two times, but failed.\nI don't think a file of 100Mb is too large to be uploaded. Is there any tips about uploading results???
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "119605": "Welcome to the 5th Facebook recruiting competition! This time, we prepared a ranking problem for you, hope you'll enjoy it.\r\n\r\nThis competition is particularly exciting for me personally - I got hired after participating in [Facebook's second recruiting challenge][1]. Now, three years later, I am helping to organize a similar challenge.\r\n\r\nWe will maintain the forum presence to answer any data- and recruiting-related questions.\r\n\r\nHave fun and thank you for participating!\r\n\r\n\r\n [1]: https://www.kaggle.com/c/facebook-ii", "119608": "What the \"opportunity at interviewing\" means? Does it mean that you can just get interview and not the job?", "119610": "Hi Competitors! Another Facebooker here to answer any questions.\r\n\r\n@Seraphimus - Correct. Unfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion.", "119613": "I understand legal reasons. But if legal conditions are met, does the winner need to pass through complete Facebook recruitment process or this competition plays the role of the recruitment process?", "119614": "I don't mean to be intentionally vague - all I can say is that it would indeed replace part of the recruiting process but not all of it. We hope that you can understand that we'd like to get the know the scientists and engineers outside the confines of Kaggle before asking them to join the team. We'd hope that you'd also like to meet us before deciding to work with us. :)", "119615": "Seraphimus,\r\n\r\nI think the description says it pretty clearly.\r\n\r\n> **PROCESS**\r\n> \r\n> Highly ranked contestants who indicate their interest will be\r\n> considered by Facebook for interviews, based on their work in the\r\n> competition and additional background.\r\n\r\nTo sum up, winning this competition is not a one-way ticket to a position at Facebook.", "119616": "There is no chance that I will ever be near the top of the leaderbord because I am a complete newbie. I was just curious. Thank you for your answers!", "119788": "[quote=Ampersand;119610]\r\nUnfortunately we cannot guarantee employment through Kaggle for many legal reasons (i.e. visa status, background check, etc.) but we'd love the opportunity to meet the winner(s) for further discussion. [/quote]\r\nDo you provide work visa support for winners or you don't?", "119878": "started another thread(https://www.kaggle.com/c/facebook-v-predicting-check-ins/forums/t/20923/candidates-must-have-or-be-eligible-to-obtain-authorization-to-work-in) before discovering this one, but i think my question(see below) is also relevant to this thread\r\n\r\nH1B quota ended in april and there would not be any new visas till the april of the next year and it is the only way for most of us to obtain authorization to work in the US, now if someone who needs H1B to legally work in the US scores hight in this competition(and meet all job requirements) does that means that facebook is agree to wait till the next year for candidate to join company or that means that he is not eligible for this competition?", "120160": "awesome dataset build.. \r\n\r\n kick ass problem.. \r\n\r\nreally hope I can make some time for this", "120300": "Has the 50% split of the test data been done randomly?", "120303": "[quote=Picarus;120300]\r\n\r\nHas the 50% split of the test data been done randomly?\r\n\r\n[/quote]\r\n\r\nShould be. From the Data page:\r\n\r\n*The train and test dataset are split based on time, and the public/private leaderboard in the test data are split randomly.*", "120306": "Did you have a particular place (e.g. downtime SFO) or a generic (kind of) place (e.g. railway platforms) in mind while designing this data set? Or are those details somehow irrelevant to this competition.", "120380": "Hi there!\r\nThere's an error in submission description\r\nhttps://www.kaggle.com/c/facebook-v-predicting-check-ins/details/evaluation\r\nFor every user check in, you must predict a space-delimited list of the businesses they check into. You may submit up to 3 predictions for each check in. The file should contain a header and have the following format:\r\nrow_id,**predictions**\r\n\r\nShould be\r\nrow_id,**place_id**\r\n\r\nThanks for an interesting competitions.", "120393": "DennisSakva, thanks! fixed!", "120394": "It was referring to the contents, not the header value. But yeah, I agree it was confusing upon first read", "120989": "Hi,\r\nI'm just curious about the \"Requirements\" for the Job:\r\n... 5 years experienced ...\r\n\r\nActually I'm doing my master thesis - I can't really have that much work experienced. :)\r\n\r\nSo, how important are these requirements?\r\n\r\nEdit: Just saw the same question&answer in another thread :)", "121499": "I have a question regarding number of predictions it is stated that: \"You may submit up to 3 predictions for each check in\". Does it mean that some of the test values have less than 3 real predictions. If yes submitting more than real predictions will impact the result.", "121500": "Each test observation has exactly 1 correct prediction: the place where the checkin effectively took place when the data generating simulation was run. You are allowed up to 3 predictions. If your first prediction is right, you get 1/1, if your 2nd prediction is right, you get 1/2, and if your 3rd prediction is right, your get 1/3. Otherwise, you get 0 for the particular observation. Hence, there is no reason *not* to make 3 predictions per test observation, in the order of likelihood.", "121547": "Andre Naef Thanks for your clear explanation.", "123427": "What kind of the rank can get the interview opportunity? Top 1 or Top 10? Thanks!", "125244": "Excuse me, Anyone has difficulty uploading the results file? \r\nI have tried two times, but failed.\r\nI don't think a file of 100Mb is too large to be uploaded. Is there any tips about uploading results???" }, "source": "meta" }