Thankyou for providing the entrance of late submission. I should trust my model and local cv instead of my intuition. \nIn my model, I split the data into 6 groups, 5 group train sets, and 1 validation set.\nIn each training set, 6 month's aggregrate feature of each id and the next 3 months' sum revenue as target.\nThe validation set is created by the same method, using 2018.02.01-2018.07.31 data to aggregrate features, using 2018.8.1 - 2018.10.15 's sum revenue as target.\nIt's difficult for me to explain my solution without code beacuse of my limited English ability ... \nThe most importance features are (aggregrated) totals.hits/totals.timeOnSite/totals.pageViews and heats which is a feature I made by myself using a linear combination of the number of hit and revenues of each month in the training set (made by pivot table). What about yours?
", "messages": [ { "id": "476394", "postDate": "02/22/2019 04:20:31", "content": "Thankyou for providing the entrance of late submission. I should trust my model and local cv instead of my intuition. \nIn my model, I split the data into 6 groups, 5 group train sets, and 1 validation set.\nIn each training set, 6 month's aggregrate feature of each id and the next 3 months' sum revenue as target.\nThe validation set is created by the same method, using 2018.02.01-2018.07.31 data to aggregrate features, using 2018.8.1 - 2018.10.15 's sum revenue as target.\nIt's difficult for me to explain my solution without code beacuse of my limited English ability ... \nThe most importance features are (aggregrated) totals.hits/totals.timeOnSite/totals.pageViews and heats which is a feature I made by myself using a linear combination of the number of hit and revenues of each month in the training set (made by pivot table). What about yours?
", "rawMarkdown": "Thankyou for providing the entrance of late submission. I should trust my model and local cv instead of my intuition. \nIn my model, I split the data into 6 groups, 5 group train sets, and 1 validation set.\nIn each training set, 6 month's aggregrate feature of each id and the next 3 months' sum revenue as target.\nThe validation set is created by the same method, using 2018.02.01-2018.07.31 data to aggregrate features, using 2018.8.1 - 2018.10.15 's sum revenue as target.\nIt's difficult for me to explain my solution without code beacuse of my limited English ability ... \nThe most importance features are (aggregrated) totals.hits/totals.timeOnSite/totals.pageViews and **heats** which is a feature I made by myself using a linear combination of the number of hit and revenues of each month in the training set (made by pivot table). What about yours?", "votes": null }, { "id": "476869", "postDate": "02/23/2019 11:04:23", "content": "I got a very bad rank ORZ.Could you share more details and your code ？honor to learn from your code
", "rawMarkdown": "I got a very bad rank ORZ.Could you share more details and your code ？honor to learn from your code", "votes": null } ], "comments": [ { "id": 476869, "author_name": "shinson", "author_url": "", "post_date": "02/23/2019 11:04:23", "content": "I got a very bad rank ORZ.Could you share more details and your code ？honor to learn from your code
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "476394": "Thankyou for providing the entrance of late submission. I should trust my model and local cv instead of my intuition. \nIn my model, I split the data into 6 groups, 5 group train sets, and 1 validation set.\nIn each training set, 6 month's aggregrate feature of each id and the next 3 months' sum revenue as target.\nThe validation set is created by the same method, using 2018.02.01-2018.07.31 data to aggregrate features, using 2018.8.1 - 2018.10.15 's sum revenue as target.\nIt's difficult for me to explain my solution without code beacuse of my limited English ability ... \nThe most importance features are (aggregrated) totals.hits/totals.timeOnSite/totals.pageViews and **heats** which is a feature I made by myself using a linear combination of the number of hit and revenues of each month in the training set (made by pivot table). What about yours?", "476869": "I got a very bad rank ORZ.Could you share more details and your code ？honor to learn from your code" }, "source": "existing_discussion", "complete": true, "competition_id": 10038, "competition_slug": "ga-customer-revenue-prediction", "team_id": 2284673, "solution_url": "/competitions/ga-customer-revenue-prediction/writeups/simon-wang-late-submission-socre-0-884-and-solutio" }