{ "id": 421036, "title": "Kaggle Winning Solutions Dataset", "url": "/competitions/2023-kaggle-ai-report/discussion/421036", "author_name": "", "post_date": "2023-07-03T16:45:14.005126900Z", "votes": 5, "comment_count": 2, "views": 0, "content": "

It's a bit late, but maybe you can still find this useful - a dataset with structured methods extracted from winning solutions write ups:

\n
    \n
  1. Dataset
    2. \n
  2. Code used to collect the dataset
    3. \n
  3. Article describing extraction process
    4. \n
  4. Example analysis
    5. \n
", "messages": [ { "id": "2328534", "postDate": "07/03/2023 16:45:14", "content": "

It's a bit late, but maybe you can still find this useful - a dataset with structured methods extracted from winning solutions write ups:

\n
    \n
  1. Dataset
    2. \n
  2. Code used to collect the dataset
    3. \n
  3. Article describing extraction process
    4. \n
  4. Example analysis
    5. \n
", "rawMarkdown": "It's a bit late, but maybe you can still find this useful - a dataset with structured methods extracted from winning solutions write ups:\n1. [Dataset](https://www.kaggle.com/datasets/thedrcat/kaggle-winning-solutions-methods)\n2. [Code](https://www.kaggle.com/code/thedrcat/using-llms-to-extract-structured-data) used to collect the dataset\n3. [Article](https://wandb.ai/darek/llmapps/reports/Using-LLMs-to-Extract-Structured-Data-OpenAI-Function-Calling-in-Action--Vmlldzo0Nzc0MzQ3) describing extraction process\n4. [Example analysis](https://www.kaggle.com/code/thedrcat/how-to-win-a-kaggle-competition)", "votes": null }, { "id": "2328546", "postDate": "07/03/2023 16:59:20", "content": "

Better late than never.
\nI hope other Kagglers could read it cause, recently, I noticed that many are just publishing Notebooks. And not reading anything else on other competitions.
\nHere, they are expected to read since they Must make at least 3 quality comments.

\n

As I wrote on your code. Your Article is Fantastic. Many kagglers should Not Miss it.

", "rawMarkdown": "Better late than never.\nI hope other Kagglers could read it cause, recently, I noticed that many are just publishing Notebooks. And not reading anything else on other competitions.\nHere, they are expected to read since they Must make at least 3 quality comments.\n\nAs I wrote on your code. Your Article is Fantastic. Many kagglers should Not Miss it.", "votes": null }, { "id": "2329033", "postDate": "07/04/2023 04:10:05", "content": "

Nice application for OpenAI function calling feature 🔥 this approach is super versatile and can be used in many scenarios!

", "rawMarkdown": "Nice application for OpenAI function calling feature 🔥 this approach is super versatile and can be used in many scenarios!", "votes": null } ], "comments": [ { "id": 2328546, "author_name": "mpwolke", "author_url": "", "post_date": "07/03/2023 16:59:20", "content": "

Better late than never.
\nI hope other Kagglers could read it cause, recently, I noticed that many are just publishing Notebooks. And not reading anything else on other competitions.
\nHere, they are expected to read since they Must make at least 3 quality comments.

\n

As I wrote on your code. Your Article is Fantastic. Many kagglers should Not Miss it.

", "votes": null, "replies": [] }, { "id": 2329033, "author_name": "conjuring92", "author_url": "", "post_date": "07/04/2023 04:10:05", "content": "

Nice application for OpenAI function calling feature 🔥 this approach is super versatile and can be used in many scenarios!

", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "2328534": "It's a bit late, but maybe you can still find this useful - a dataset with structured methods extracted from winning solutions write ups:\n1. [Dataset](https://www.kaggle.com/datasets/thedrcat/kaggle-winning-solutions-methods)\n2. [Code](https://www.kaggle.com/code/thedrcat/using-llms-to-extract-structured-data) used to collect the dataset\n3. [Article](https://wandb.ai/darek/llmapps/reports/Using-LLMs-to-Extract-Structured-Data-OpenAI-Function-Calling-in-Action--Vmlldzo0Nzc0MzQ3) describing extraction process\n4. [Example analysis](https://www.kaggle.com/code/thedrcat/how-to-win-a-kaggle-competition)", "2328546": "Better late than never.\nI hope other Kagglers could read it cause, recently, I noticed that many are just publishing Notebooks. And not reading anything else on other competitions.\nHere, they are expected to read since they Must make at least 3 quality comments.\n\nAs I wrote on your code. Your Article is Fantastic. Many kagglers should Not Miss it.", "2329033": "Nice application for OpenAI function calling feature 🔥 this approach is super versatile and can be used in many scenarios!" }, "source": "meta" }