![\"output-format\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F1441360%2F00ba9f7b4a4b3003187e46550233f61b%2Fconfusion.png?generation=1594430413627480&alt=media\")
In the Overview section its said:
\n\nHowever regarding to output format the Data section says:
\n\nSo 16 columns are expected but what exactly does each columns represent? It's not clear to me yet. (should I predict for the day of 'today\" also therefore summing to 8 days?).
\n\nFinally, the outrput format made me confuse:\n
What is a \"combination of days shifts/ status transition\"?\nFinally, the final submission format of that notebook seems to me different from the pattern described.
\n\nThanks in advance.
", "messages": [ { "id": 924518, "postDate": "2020-07-11T13:37:53.167Z", "content": "Hi maffei2443,
\n\nYou need to predict 112 variables per date. It's 112 = 7 (Days ahead) * 4 (shifts) * 2 (states) * 2 (transactions - entering or leaving).
\n\nThe output format in the submission file needs to be:\n <id>_output_<k>_<d>
Where, id is a register ID.. K is combination that shifts, states and transactions (entering or leavng) and could be 1 to 16 and d is days ahead and could be 0 to 6
", "rawMarkdown": "Hi maffei2443,\n\nYou need to predict 112 variables per date. It's 112 = 7 (Days ahead) * 4 (shifts) * 2 (states) * **2** (transactions - entering or leaving).\n\nThe output format in the submission file needs to be:\n `", "votes": 3, "replies": [ { "id": 926642, "postDate": "2020-07-12T20:17:38.170Z", "content": "Thanks for replying.\nHowever I don't understand yet:\n 1. what is a \"shift\"? Are they represented by the contiguous dark green arrows?
\n\nI understand that for each day there must be 4 * 2 * 2 = 16 columns, eacho one a combination?. Still, I do not unserstand the meaning of the output format, specifically the combination of day/status transitions. What would be an example of \"combination\"?
", "rawMarkdown": "Thanks for replying.\nHowever I don't understand yet:\n 1. what is a \"shift\"? Are they represented by the contiguous dark green arrows?\n\n2. Are the \"states\" you refer to \"available\" and \"unavailable\"? \n\nI understand that for each day there must be 4 * 2 * 2 = **16 columns**, eacho one a combination?. Still, I do not unserstand the meaning of the output format, **specifically the combination of day/status transitions**. What would be an example of \"combination\"?", "votes": 1 }, { "id": 926739, "postDate": "2020-07-13T00:30:46.937Z", "content": "> what is a \"shift\"? Are they represented by the contiguous dark green arrows?
\n\nDay \"shifts\" are \"morning\", \"evening\", \"afternoon\" and \"night\"
\n\n> Are the \"states\" you refer to \"available\" and \"unavailable\"?
\n\nyes
\n\n> What would be an example of \"combination\"?
\n\nEach day, we have 16 combinations of values... because, we have 4 shifts [morning, evening, afternoon, night], 2 states [available, unavailable] and 2 transitions [entering , leaving].. so, I think like this:
\n\nOBS: This explanation is my opinion because in the original text does not explain what is \"Combination of days/status transitions\"
", "rawMarkdown": "> what is a \"shift\"? Are they represented by the contiguous dark green arrows?\n\nDay \"shifts\" are \"morning\", \"evening\", \"afternoon\" and \"night\"\n\n> Are the \"states\" you refer to \"available\" and \"unavailable\"?\n\nyes\n\n> What would be an example of \"combination\"?\n\nEach day, we have 16 combinations of values... because, we have 4 shifts [morning, evening, afternoon, night], 2 states [available, unavailable] and 2 transitions [entering , leaving].. so, I think like this:\n\n* output\\_1\\_1 = **Entering** **available** in the **morning**\n* output\\_2\\_1 = **Leaving** **available** in the **morning**\n* output\\_3\\_1 = **Entering** **unavailable** in the **morning**\n* output\\_4\\_1 = **Leaving** **unavailable** in the **morning**\n* output\\_5\\_1 = **Entering** **available** in the **evening**\n* output\\_6\\_1 = **Leaving** **available** in the **evening**\n* output\\_7\\_1 = **Entering** **unavailable** in the **evening**\n* output\\_8\\_1 = **Leaving** **unavailable** in the **evening**\n....\n\n\n\nOBS: This explanation is my opinion because in the original text does not explain what is \"Combination of days/status transitions\"\n", "votes": 3 } ] }, { "id": 923566, "postDate": "2020-07-11T01:29:24.217Z", "content": "In the Overview section its said:
\n\nHowever regarding to output format the Data section says:
\n\nSo 16 columns are expected but what exactly does each columns represent? It's not clear to me yet. (should I predict for the day of 'today\" also therefore summing to 8 days?).
\n\nFinally, the outrput format made me confuse:\n
What is a \"combination of days shifts/ status transition\"?\nFinally, the final submission format of that notebook seems to me different from the pattern described.
\n\nThanks in advance.
", "rawMarkdown": "In the *Overview* section its said: \n\n### The goal is to predict the number of cars entering and leaving each of the two status, for each of the four shifts in a day, for one week ahead.\n\nHowever regarding to output format the *Data* section says:\n\n### The output variables consist of a set of 16 (sixteen) columns, representing each of the 7 (seven) days ahead of the tuple date, 4 (four) shifts a day and 4 (four) status transitions a day, two for each status (entering or leaving)\n\nSo 16 columns are expected but what exactly does each columns represent? It's not clear to me yet. (should I predict for the day of 'today\" also therefore summing to 8 days?).\n\nFinally, the outrput format made me confuse:\n \n\nWhat is a \"**combination of days shifts/ status transition**\"?\nFinally, the final submission format of [that notebook](https://www.kaggle.com/marlesson/simple-pytorch-model) seems to me different from the pattern described.\n\nThanks in advance.", "votes": 4 }, { "id": 929476, "postDate": "2020-07-14T17:35:05.273Z", "content": "I will check the information with the data owners, but I think Marlesson is right in his interpretation.
", "rawMarkdown": "I will check the information with the data owners, but I think Marlesson is right in his interpretation.", "votes": 1 }, { "id": 977248, "postDate": "2020-08-19T11:24:43.467Z", "content": "The data owners have confirmed that Marlesson's interpretation is right.
", "rawMarkdown": "The data owners have confirmed that Marlesson's interpretation is right." }, { "id": 973969, "postDate": "2020-08-17T16:29:20.303Z", "rawMarkdown": "", "votes": 1, "isDeleted": true } ], "comments": [ { "id": 924518, "author_name": "Marlesson", "author_url": "", "post_date": "2020-07-11T13:37:53.167000", "content": "Hi maffei2443,
\n\nYou need to predict 112 variables per date. It's 112 = 7 (Days ahead) * 4 (shifts) * 2 (states) * 2 (transactions - entering or leaving).
\n\nThe output format in the submission file needs to be:\n <id>_output_<k>_<d>
Where, id is a register ID.. K is combination that shifts, states and transactions (entering or leavng) and could be 1 to 16 and d is days ahead and could be 0 to 6
", "votes": 3, "replies": [ { "id": 926642, "author_name": "Leonardo Maffei da Silva", "author_url": "", "post_date": "2020-07-12T20:17:38.170000", "content": "Thanks for replying.\nHowever I don't understand yet:\n 1. what is a \"shift\"? Are they represented by the contiguous dark green arrows?
\n\nI understand that for each day there must be 4 * 2 * 2 = 16 columns, eacho one a combination?. Still, I do not unserstand the meaning of the output format, specifically the combination of day/status transitions. What would be an example of \"combination\"?
", "votes": 1, "replies": [] }, { "id": 926739, "author_name": "Marlesson", "author_url": "", "post_date": "2020-07-13T00:30:46.937000", "content": "> what is a \"shift\"? Are they represented by the contiguous dark green arrows?
\n\nDay \"shifts\" are \"morning\", \"evening\", \"afternoon\" and \"night\"
\n\n> Are the \"states\" you refer to \"available\" and \"unavailable\"?
\n\nyes
\n\n> What would be an example of \"combination\"?
\n\nEach day, we have 16 combinations of values... because, we have 4 shifts [morning, evening, afternoon, night], 2 states [available, unavailable] and 2 transitions [entering , leaving].. so, I think like this:
\n\nOBS: This explanation is my opinion because in the original text does not explain what is \"Combination of days/status transitions\"
", "votes": 3, "replies": [] } ] }, { "id": 929476, "author_name": "AnaCarolinaLorena", "author_url": "", "post_date": "2020-07-14T17:35:05.273000", "content": "I will check the information with the data owners, but I think Marlesson is right in his interpretation.
", "votes": 1, "replies": [] }, { "id": 977248, "author_name": "AnaCarolinaLorena", "author_url": "", "post_date": "2020-08-19T11:24:43.467000", "content": "The data owners have confirmed that Marlesson's interpretation is right.
", "votes": 0, "replies": [] }, { "id": 973969, "author_name": "", "author_url": "", "post_date": "2020-08-17T16:29:20.303000", "content": "", "votes": 1, "replies": [] } ], "raw_markdown_by_id": { "924518": "Hi maffei2443,\n\nYou need to predict 112 variables per date. It's 112 = 7 (Days ahead) * 4 (shifts) * 2 (states) * **2** (transactions - entering or leaving).\n\nThe output format in the submission file needs to be:\n `", "923566": "In the *Overview* section its said: \n\n### The goal is to predict the number of cars entering and leaving each of the two status, for each of the four shifts in a day, for one week ahead.\n\nHowever regarding to output format the *Data* section says:\n\n### The output variables consist of a set of 16 (sixteen) columns, representing each of the 7 (seven) days ahead of the tuple date, 4 (four) shifts a day and 4 (four) status transitions a day, two for each status (entering or leaving)\n\nSo 16 columns are expected but what exactly does each columns represent? It's not clear to me yet. (should I predict for the day of 'today\" also therefore summing to 8 days?).\n\nFinally, the outrput format made me confuse:\n \n\nWhat is a \"**combination of days shifts/ status transition**\"?\nFinally, the final submission format of [that notebook](https://www.kaggle.com/marlesson/simple-pytorch-model) seems to me different from the pattern described.\n\nThanks in advance.", "929476": "I will check the information with the data owners, but I think Marlesson is right in his interpretation.", "977248": "The data owners have confirmed that Marlesson's interpretation is right.", "973969": "" } }