{ "id": 328651, "title": "Pivot datasets - One row for customer_ID", "url": "/competitions/amex-default-prediction/discussion/328651", "author_name": "", "post_date": "2022-06-02T10:11:46.463921500Z", "votes": 11, "comment_count": 5, "views": 0, "content": "

I have created feather train & test files with pivot information, that is, one row for each customer_ID and the transposition for each feature:

\n

XXX_0 is the most recent information and XXX_12 is the oldest (not all customers necessarily have 13 time points, in this cases the information is filled with missing values)

\n

All features are in float16 or category format.

\n

Link to Dataset - https://www.kaggle.com/datasets/jordimoragas/amex-pivot-dataset-feather

", "messages": [ { "id": "1809004", "postDate": "06/02/2022 10:11:46", "content": "

I have created feather train & test files with pivot information, that is, one row for each customer_ID and the transposition for each feature:

\n

XXX_0 is the most recent information and XXX_12 is the oldest (not all customers necessarily have 13 time points, in this cases the information is filled with missing values)

\n

All features are in float16 or category format.

\n

Link to Dataset - https://www.kaggle.com/datasets/jordimoragas/amex-pivot-dataset-feather

", "rawMarkdown": "I have created feather train & test files with pivot information, that is, one row for each customer_ID and the transposition for each feature:\n\nXXX_0 is the most recent information and XXX_12 is the oldest (not all customers necessarily have 13 time points, in this cases the information is filled with missing values)\n\nAll features are in float16 or category format.\n\nLink to Dataset - https://www.kaggle.com/datasets/jordimoragas/amex-pivot-dataset-feather", "votes": null }, { "id": "1810199", "postDate": "06/03/2022 10:41:42", "content": "

Hello again. Created aggregate datasets: One row for each customer_ID with new features like last observation, minimum, maximum, mean, deviation… See the dataset page for more details: https://www.kaggle.com/datasets/jordimoragas/amex-agg-dataset-feather

\n

Very low memory consuming!

", "rawMarkdown": "Hello again. Created aggregate datasets: One row for each customer_ID with new features like last observation, minimum, maximum, mean, deviation... See the dataset page for more details: https://www.kaggle.com/datasets/jordimoragas/amex-agg-dataset-feather\n\nVery low memory consuming!", "votes": null }, { "id": "1816995", "postDate": "06/10/2022 18:28:48", "content": "

Very helpful. Thanks a lot

", "rawMarkdown": "Very helpful. Thanks a lot", "votes": null }, { "id": "1881534", "postDate": "08/02/2022 15:30:49", "content": "

can you share the code for generating these files

", "rawMarkdown": "can you share the code for generating these files", "votes": null }, { "id": "1908093", "postDate": "08/21/2022 11:12:20", "content": "

Kindly upload the code you used to create these datasets to better understand the logic.

", "rawMarkdown": "Kindly upload the code you used to create these datasets to better understand the logic.", "votes": null }, { "id": "1908437", "postDate": "08/21/2022 17:10:37", "content": "

How did you treat customer_ID's with less than 13 statements in your database?

", "rawMarkdown": "How did you treat customer_ID's with less than 13 statements in your database?", "votes": null } ], "comments": [ { "id": 1810199, "author_name": "jordimoragas", "author_url": "", "post_date": "06/03/2022 10:41:42", "content": "

Hello again. Created aggregate datasets: One row for each customer_ID with new features like last observation, minimum, maximum, mean, deviation… See the dataset page for more details: https://www.kaggle.com/datasets/jordimoragas/amex-agg-dataset-feather

\n

Very low memory consuming!

", "votes": null, "replies": [ { "id": 1881534, "author_name": "rajesh372", "author_url": "", "post_date": "08/02/2022 15:30:49", "content": "

can you share the code for generating these files

", "votes": null, "replies": [] } ] }, { "id": 1816995, "author_name": "aninda", "author_url": "", "post_date": "06/10/2022 18:28:48", "content": "

Very helpful. Thanks a lot

", "votes": null, "replies": [] }, { "id": 1908093, "author_name": "ydvaakash", "author_url": "", "post_date": "08/21/2022 11:12:20", "content": "

Kindly upload the code you used to create these datasets to better understand the logic.

", "votes": null, "replies": [] }, { "id": 1908437, "author_name": "rasoulmojtahedzadeh", "author_url": "", "post_date": "08/21/2022 17:10:37", "content": "

How did you treat customer_ID's with less than 13 statements in your database?

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1809004": "I have created feather train & test files with pivot information, that is, one row for each customer_ID and the transposition for each feature:\n\nXXX_0 is the most recent information and XXX_12 is the oldest (not all customers necessarily have 13 time points, in this cases the information is filled with missing values)\n\nAll features are in float16 or category format.\n\nLink to Dataset - https://www.kaggle.com/datasets/jordimoragas/amex-pivot-dataset-feather", "1810199": "Hello again. Created aggregate datasets: One row for each customer_ID with new features like last observation, minimum, maximum, mean, deviation... See the dataset page for more details: https://www.kaggle.com/datasets/jordimoragas/amex-agg-dataset-feather\n\nVery low memory consuming!", "1816995": "Very helpful. Thanks a lot", "1881534": "can you share the code for generating these files", "1908093": "Kindly upload the code you used to create these datasets to better understand the logic.", "1908437": "How did you treat customer_ID's with less than 13 statements in your database?" }, "source": "meta" }