{ "id": 518948, "title": "Leaderboard Dynamics Visualisation | Locate your team on the plot", "url": "/competitions/leash-BELKA/discussion/518948", "author_name": "Samvel Kocharyan", "post_date": "2024-07-09T01:36:54.380000", "votes": 11, "comment_count": 4, "views": 0, "content": "

Here is NeurIPS 2024 - Predict New Medicines with BELKA Leaderboard and shakeup visualisation plot. Full interactive version could be found in public notebook. There you could locate your team on the plot (Copy & Edit). Have fun!

\n

\n

More info: Churn analysis by @ravi20076

", "messages": [ { "id": 2912598, "postDate": "2024-07-09T01:36:54.380Z", "content": "

Here is NeurIPS 2024 - Predict New Medicines with BELKA Leaderboard and shakeup visualisation plot. Full interactive version could be found in public notebook. There you could locate your team on the plot (Copy & Edit). Have fun!

\n

\n

More info: Churn analysis by @ravi20076

", "rawMarkdown": "Here is NeurIPS 2024 - Predict New Medicines with BELKA Leaderboard and shakeup visualisation plot. Full interactive version could be found [in public notebook](https://www.kaggle.com/code/samvelkoch/belka-interactive-leaderboard-and-shake). There you could locate your team on the plot (Copy & Edit). Have fun!\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F10356799%2Fc8a0374df03cb813828c969c5285bce4%2FScreenshot%202024-07-09%20at%203.31.45.png?generation=1720488864863046&alt=media)\n\nMore info: [Churn analysis](https://www.kaggle.com/competitions/leash-BELKA/discussion/518942) by @ravi20076", "votes": 11 }, { "id": 2912749, "postDate": "2024-07-09T04:02:03.143Z", "content": "

Thanks for sharing a nice plot. It looks like a sprinkler!

", "rawMarkdown": "Thanks for sharing a nice plot. It looks like a sprinkler!", "votes": 1 }, { "id": 2912672, "postDate": "2024-07-09T02:53:40.880Z", "rawMarkdown": "", "votes": 1, "isDeleted": true, "replies": [ { "id": 2912681, "postDate": "2024-07-09T03:05:22.450Z", "content": "

Bio problems are just harder. No human can do them, but we often have good intuitions in other domains, like language, images, and time-series. BELKA was especially hard because 100m is a minuscule fraction of the 10^60 possible molecules, and there is no natural manifold to restrict the problem to.

", "rawMarkdown": "Bio problems are just harder. No human can do them, but we often have good intuitions in other domains, like language, images, and time-series. BELKA was especially hard because 100m is a minuscule fraction of the 10^60 possible molecules, and there is no natural manifold to restrict the problem to.", "votes": 1 }, { "id": 2913303, "postDate": "2024-07-09T11:52:02.270Z", "content": "

3 options:

\n
    \n
  1. My Dataset. Created in 1st hour after competition final
    2. \n
  2. Leaderboard Links (look for the \"Raw Data\" button on the right in Desktop Version of the Leaderboards)
    3. \n
  3. Kaggle Meta Dataset. Finalized data. Available next few days after the competition final.
    4. \n
", "rawMarkdown": "3 options:\n1. [My Dataset](https://www.kaggle.com/datasets/samvelkoch/belka-leaderboard). Created in 1st hour after competition final\n2. [Leaderboard Links](https://www.kaggle.com/competitions/leash-BELKA/leaderboard) (look for the \"Raw Data\" button on the right in Desktop Version of the Leaderboards)\n3. [Kaggle Meta Dataset](https://www.kaggle.com/datasets/kaggle/meta-kaggle/data). Finalized data. Available next few days after the competition final. " } ] } ], "comments": [ { "id": 2912749, "author_name": "C R Suthikshn Kumar", "author_url": "", "post_date": "2024-07-09T04:02:03.143000", "content": "

Thanks for sharing a nice plot. It looks like a sprinkler!

", "votes": 1, "replies": [] }, { "id": 2912672, "author_name": "", "author_url": "", "post_date": "2024-07-09T02:53:40.880000", "content": "", "votes": 1, "replies": [ { "id": 2912681, "author_name": "Andrew D. Blevins", "author_url": "", "post_date": "2024-07-09T03:05:22.450000", "content": "

Bio problems are just harder. No human can do them, but we often have good intuitions in other domains, like language, images, and time-series. BELKA was especially hard because 100m is a minuscule fraction of the 10^60 possible molecules, and there is no natural manifold to restrict the problem to.

", "votes": 1, "replies": [] }, { "id": 2913303, "author_name": "Samvel Kocharyan", "author_url": "", "post_date": "2024-07-09T11:52:02.270000", "content": "

3 options:

\n
    \n
  1. My Dataset. Created in 1st hour after competition final
    2. \n
  2. Leaderboard Links (look for the \"Raw Data\" button on the right in Desktop Version of the Leaderboards)
    3. \n
  3. Kaggle Meta Dataset. Finalized data. Available next few days after the competition final.
    4. \n
", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "2912598": "Here is NeurIPS 2024 - Predict New Medicines with BELKA Leaderboard and shakeup visualisation plot. Full interactive version could be found [in public notebook](https://www.kaggle.com/code/samvelkoch/belka-interactive-leaderboard-and-shake). There you could locate your team on the plot (Copy & Edit). Have fun!\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F10356799%2Fc8a0374df03cb813828c969c5285bce4%2FScreenshot%202024-07-09%20at%203.31.45.png?generation=1720488864863046&alt=media)\n\nMore info: [Churn analysis](https://www.kaggle.com/competitions/leash-BELKA/discussion/518942) by @ravi20076", "2912749": "Thanks for sharing a nice plot. It looks like a sprinkler!", "2912672": "" } }