{ "id": 390580, "title": "Jo Wilder Game Play EDA", "url": "/competitions/predict-student-performance-from-game-play/discussion/390580", "author_name": "", "post_date": "2023-02-26T09:50:15.100204100Z", "votes": 14, "comment_count": 5, "views": 0, "content": "

It looks like most of the publicly available kernels are model baselines. So, I created an exploratory data analysis with a general overview of the data.

\n

Some preview:
\n\"\"
\n\"\"
\n\"\"

", "messages": [ { "id": "2160015", "postDate": "02/26/2023 09:50:15", "content": "

It looks like most of the publicly available kernels are model baselines. So, I created an exploratory data analysis with a general overview of the data.

\n

Some preview:
\n\"\"
\n\"\"
\n\"\"

", "rawMarkdown": "It looks like most of the publicly available kernels are model baselines. So, I created an [exploratory data analysis](https://www.kaggle.com/code/demche/student-performance-from-game-play-eda) with a general overview of the data.\n\nSome preview:\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5196404%2F6e3664ab267641afae07058cd6b839cc%2F1.PNG?generation=1677404967224776&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5196404%2F30756dba77dd1fda40c2dbb587758f4f%2F4.PNG?generation=1677404975968071&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5196404%2Fe43438cd33dcea2d9860961a2a67dccf%2F2.PNG?generation=1677404984537317&alt=media)", "votes": null }, { "id": "2160149", "postDate": "02/26/2023 12:28:07", "content": "

excellent EDA, learned a lot and I spent some time doing EDA myself.

", "rawMarkdown": "excellent EDA, learned a lot and I spent some time doing EDA myself.", "votes": null }, { "id": "2160172", "postDate": "02/26/2023 12:50:38", "content": "

I have a hypothesis on the unknown_part column:

\n
count    \nmean        \nstd         \n          \n%         \n%         \n%         \n         \nName: unknown_part, dtype: float6\n
\n

Doesn't that look like session time in minutes ? I know there is already the elasped_time column where we can take the max, but it proved to be rather not trust-worthy. Min being 0 is wierd but again it's a hypothesis

", "rawMarkdown": "I have a hypothesis on the unknown_part column: \n```python\ncount 11779.000000\nmean 47.731217\nstd 28.745162\nmin 0.000000\n25% 24.000000\n50% 50.000000\n75% 70.000000\nmax 96.000000\nName: unknown_part, dtype: float6\n```\n\nDoesn't that look like **session time** in minutes ? I know there is already the elasped_time column where we can take the max, but it proved to be rather not trust-worthy. Min being 0 is wierd but again it's a hypothesis", "votes": null }, { "id": "2160635", "postDate": "02/26/2023 20:13:06", "content": "

I answered here. In short, I believe the unknown part is the players' counter: 0 for the first player with the specific (non-unique) date/time combination, 1 for the second, etc

", "rawMarkdown": "I answered [here](https://www.kaggle.com/code/demche/student-performance-from-game-play-eda/comments#2160632). In short, I believe the unknown part is the players' counter: 0 for the first player with the specific (non-unique) date/time combination, 1 for the second, etc", "votes": null }, { "id": "2221592", "postDate": "04/14/2023 11:51:10", "content": "

Very wonderful. Did you write this table in python?@demche

", "rawMarkdown": "Very wonderful. Did you write this table in python?@demche", "votes": null }, { "id": "2221953", "postDate": "04/14/2023 18:08:11", "content": "

Not sure which of the tables do you mean. Click Show hidden code in the notebook to see code used to create specific table/diagram

", "rawMarkdown": "Not sure which of the tables do you mean. Click `Show hidden code` in the notebook to see code used to create specific table/diagram", "votes": null } ], "comments": [ { "id": 2160149, "author_name": "janmpia", "author_url": "", "post_date": "02/26/2023 12:28:07", "content": "

excellent EDA, learned a lot and I spent some time doing EDA myself.

", "votes": null, "replies": [] }, { "id": 2160172, "author_name": "janmpia", "author_url": "", "post_date": "02/26/2023 12:50:38", "content": "

I have a hypothesis on the unknown_part column:

\n
count    \nmean        \nstd         \n          \n%         \n%         \n%         \n         \nName: unknown_part, dtype: float6\n
\n

Doesn't that look like session time in minutes ? I know there is already the elasped_time column where we can take the max, but it proved to be rather not trust-worthy. Min being 0 is wierd but again it's a hypothesis

", "votes": null, "replies": [ { "id": 2160635, "author_name": "demche", "author_url": "", "post_date": "02/26/2023 20:13:06", "content": "

I answered here. In short, I believe the unknown part is the players' counter: 0 for the first player with the specific (non-unique) date/time combination, 1 for the second, etc

", "votes": null, "replies": [] } ] }, { "id": 2221592, "author_name": "risakashiwabara", "author_url": "", "post_date": "04/14/2023 11:51:10", "content": "

Very wonderful. Did you write this table in python?@demche

", "votes": null, "replies": [ { "id": 2221953, "author_name": "demche", "author_url": "", "post_date": "04/14/2023 18:08:11", "content": "

Not sure which of the tables do you mean. Click Show hidden code in the notebook to see code used to create specific table/diagram

", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "2160015": "It looks like most of the publicly available kernels are model baselines. So, I created an [exploratory data analysis](https://www.kaggle.com/code/demche/student-performance-from-game-play-eda) with a general overview of the data.\n\nSome preview:\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5196404%2F6e3664ab267641afae07058cd6b839cc%2F1.PNG?generation=1677404967224776&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5196404%2F30756dba77dd1fda40c2dbb587758f4f%2F4.PNG?generation=1677404975968071&alt=media)\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F5196404%2Fe43438cd33dcea2d9860961a2a67dccf%2F2.PNG?generation=1677404984537317&alt=media)", "2160149": "excellent EDA, learned a lot and I spent some time doing EDA myself.", "2160172": "I have a hypothesis on the unknown_part column: \n```python\ncount 11779.000000\nmean 47.731217\nstd 28.745162\nmin 0.000000\n25% 24.000000\n50% 50.000000\n75% 70.000000\nmax 96.000000\nName: unknown_part, dtype: float6\n```\n\nDoesn't that look like **session time** in minutes ? I know there is already the elasped_time column where we can take the max, but it proved to be rather not trust-worthy. Min being 0 is wierd but again it's a hypothesis", "2160635": "I answered [here](https://www.kaggle.com/code/demche/student-performance-from-game-play-eda/comments#2160632). In short, I believe the unknown part is the players' counter: 0 for the first player with the specific (non-unique) date/time combination, 1 for the second, etc", "2221592": "Very wonderful. Did you write this table in python?@demche", "2221953": "Not sure which of the tables do you mean. Click `Show hidden code` in the notebook to see code used to create specific table/diagram" }, "source": "meta" }