Hi Everyone, \nI am new to kaggle, So if my questions are silly please forgive me for that. So while exploring the kernals by different participant I saw that the train and test data consists of 30 columns but when i read the files it has only 12 columns. Can anyone tell why do I have less columns?\nThank you so much.
", "messages": [ { "id": 398152, "postDate": "2018-10-03T15:31:03.683Z", "content": "Under the Data Description section of the Competition page (link), the following data fields are listed: \nfullVisitorId, channelGrouping, date, device, geoNetwork, sessionId, socialEngagementType, totals, trafficSource, visitId, visitNumber, visitStartTime
\n\nAs you can see, the competition providers (Google) have specified that there are only 12 data fields (and thus, there should be only 12 columns in the data). I'm not sure why there would be 30-column datasets in someone's kernel, but if you can reply with a link to the kernel I'm sure we could figure it out.
", "rawMarkdown": "Under the Data Description section of the Competition page ([link][1]), the following data fields are listed: \nfullVisitorId, channelGrouping, date, device, geoNetwork, sessionId, socialEngagementType, totals, trafficSource, visitId, visitNumber, visitStartTime\n\nAs you can see, the competition providers (Google) have specified that there are only 12 data fields (and thus, there should be only 12 columns in the data). I'm not sure why there would be 30-column datasets in someone's kernel, but if you can reply with a link to the kernel I'm sure we could figure it out. \n\n [1]: https://www.kaggle.com/c/ga-customer-revenue-prediction/data", "votes": 2, "replies": [ { "id": 398176, "postDate": "2018-10-03T16:14:35.220Z", "content": "Some columns are in JSON format. For example, in totals columns, you will see that there are bounces, hits, page views, transaction revenue. So when making separate columns for each the variable you will get 30 columns.
\n\nSo the kernels you are seeing are just breaking the JSON columns to separate columns
\n\nHope this clarifies your doubt
", "rawMarkdown": "Some columns are in JSON format. For example, in totals columns, you will see that there are bounces, hits, page views, transaction revenue. So when making separate columns for each the variable you will get 30 columns.\n\n\nSo the kernels you are seeing are just breaking the JSON columns to separate columns\n\n\nHope this clarifies your doubt", "votes": 3 }, { "id": 398527, "postDate": "2018-10-04T08:25:45.030Z", "content": "Thank you for taking out time to clear my doubt. I need an another help, can you please tell me how to make separate Columns. Thank you so much.
", "rawMarkdown": "Thank you for taking out time to clear my doubt. I need an another help, can you please tell me how to make separate Columns. Thank you so much." }, { "id": 398529, "postDate": "2018-10-04T08:28:41.197Z", "rawMarkdown": "", "isDeleted": true }, { "id": 398530, "postDate": "2018-10-04T08:29:12.170Z", "content": "Thank you for taking out time to clear my doubts. Here is the link to the [kernal] (https://www.kaggle.com/ashishpatel26/bayesian-lightgbm-fe-groupkfold-cv) . If you could help me out with this then that would be wonderful. Thank you
", "rawMarkdown": "Thank you for taking out time to clear my doubts. Here is the link to the [kernal] (https://www.kaggle.com/ashishpatel26/bayesian-lightgbm-fe-groupkfold-cv) . If you could help me out with this then that would be wonderful. Thank you" } ] }, { "id": 398137, "postDate": "2018-10-03T15:00:24.063Z", "content": "Hi Everyone, \nI am new to kaggle, So if my questions are silly please forgive me for that. So while exploring the kernals by different participant I saw that the train and test data consists of 30 columns but when i read the files it has only 12 columns. Can anyone tell why do I have less columns?\nThank you so much.
", "rawMarkdown": "Hi Everyone, \nI am new to kaggle, So if my questions are silly please forgive me for that. So while exploring the kernals by different participant I saw that the train and test data consists of 30 columns but when i read the files it has only 12 columns. Can anyone tell why do I have less columns?\nThank you so much." } ], "comments": [ { "id": 398152, "author_name": "Alec Kulakowski", "author_url": "", "post_date": "2018-10-03T15:31:03.683000", "content": "Under the Data Description section of the Competition page (link), the following data fields are listed: \nfullVisitorId, channelGrouping, date, device, geoNetwork, sessionId, socialEngagementType, totals, trafficSource, visitId, visitNumber, visitStartTime
\n\nAs you can see, the competition providers (Google) have specified that there are only 12 data fields (and thus, there should be only 12 columns in the data). I'm not sure why there would be 30-column datasets in someone's kernel, but if you can reply with a link to the kernel I'm sure we could figure it out.
", "votes": 2, "replies": [ { "id": 398176, "author_name": "redshellspy", "author_url": "", "post_date": "2018-10-03T16:14:35.220000", "content": "Some columns are in JSON format. For example, in totals columns, you will see that there are bounces, hits, page views, transaction revenue. So when making separate columns for each the variable you will get 30 columns.
\n\nSo the kernels you are seeing are just breaking the JSON columns to separate columns
\n\nHope this clarifies your doubt
", "votes": 3, "replies": [] }, { "id": 398527, "author_name": "Souvik Banerjee", "author_url": "", "post_date": "2018-10-04T08:25:45.030000", "content": "Thank you for taking out time to clear my doubt. I need an another help, can you please tell me how to make separate Columns. Thank you so much.
", "votes": 0, "replies": [] }, { "id": 398529, "author_name": "", "author_url": "", "post_date": "2018-10-04T08:28:41.197000", "content": "", "votes": 0, "replies": [] }, { "id": 398530, "author_name": "Souvik Banerjee", "author_url": "", "post_date": "2018-10-04T08:29:12.170000", "content": "Thank you for taking out time to clear my doubts. Here is the link to the [kernal] (https://www.kaggle.com/ashishpatel26/bayesian-lightgbm-fe-groupkfold-cv) . If you could help me out with this then that would be wonderful. Thank you
", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "398152": "Under the Data Description section of the Competition page ([link][1]), the following data fields are listed: \nfullVisitorId, channelGrouping, date, device, geoNetwork, sessionId, socialEngagementType, totals, trafficSource, visitId, visitNumber, visitStartTime\n\nAs you can see, the competition providers (Google) have specified that there are only 12 data fields (and thus, there should be only 12 columns in the data). I'm not sure why there would be 30-column datasets in someone's kernel, but if you can reply with a link to the kernel I'm sure we could figure it out. \n\n [1]: https://www.kaggle.com/c/ga-customer-revenue-prediction/data", "398137": "Hi Everyone, \nI am new to kaggle, So if my questions are silly please forgive me for that. So while exploring the kernals by different participant I saw that the train and test data consists of 30 columns but when i read the files it has only 12 columns. Can anyone tell why do I have less columns?\nThank you so much." } }