{
  "id": 46838,
  "title": "Looking for teammate",
  "url": "/competitions/favorita-grocery-sales-forecasting/discussion/46838",
  "author_name": "Jacques Peeters",
  "post_date": "2018-01-03T23:10:43.398000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hello everyone,</p>\n\n<p>I'm looking for a last minute teammate(s) for this competition. </p>\n\n<h3>Which language am I using?</h3>\n\n<p>My code base is in R for this competition. However I know how to code on Python too. See my Python solution for Instacart where I finished 52nd / top2 % \n<a href=\"https://github.com/jacquespeeters/instacart-market-basket-analysis\">https://github.com/jacquespeeters/instacart-market-basket-analysis</a></p>\n\n<h3>What is my workspace setup?</h3>\n\n<p>I'm prototyping on my laptop when I'm training on hundreds thousands of lines. On millions lines I'm working on Google Cloud Platform. </p>\n\n<h3>What I have done on this competition?</h3>\n\n<p>I had a close look at Ceshine kernels. However I reimplemented it enterely by myself in R with mostly dplyr style in order to be flexible and readable. Only focused on feature engineering, no model parameter tuning. However my solution isn't much better than public kernels, (0.53) and I'm quite disapointed. I'm wondering if I have a problem in my code which I don't spot. </p>\n\n<h3>What am I looking for?</h3>\n\n<ul>\n<li>Code review</li>\n<li>Ideas sharing</li>\n</ul>\n\n<h3>Who am I looking for?</h3>\n\n<ul>\n<li>Better score than public kernels</li>\n<li>Or previous competitions medals</li>\n<li>Ability to read R code </li>\n</ul>\n\n<p>Jacques,</p>",
  "messages": [
    {
      "id": 264813,
      "postDate": "2018-01-03T23:10:43.400Z",
      "content": "<p>Hello everyone,</p>\n\n<p>I'm looking for a last minute teammate(s) for this competition. </p>\n\n<h3>Which language am I using?</h3>\n\n<p>My code base is in R for this competition. However I know how to code on Python too. See my Python solution for Instacart where I finished 52nd / top2 % \n<a href=\"https://github.com/jacquespeeters/instacart-market-basket-analysis\">https://github.com/jacquespeeters/instacart-market-basket-analysis</a></p>\n\n<h3>What is my workspace setup?</h3>\n\n<p>I'm prototyping on my laptop when I'm training on hundreds thousands of lines. On millions lines I'm working on Google Cloud Platform. </p>\n\n<h3>What I have done on this competition?</h3>\n\n<p>I had a close look at Ceshine kernels. However I reimplemented it enterely by myself in R with mostly dplyr style in order to be flexible and readable. Only focused on feature engineering, no model parameter tuning. However my solution isn't much better than public kernels, (0.53) and I'm quite disapointed. I'm wondering if I have a problem in my code which I don't spot. </p>\n\n<h3>What am I looking for?</h3>\n\n<ul>\n<li>Code review</li>\n<li>Ideas sharing</li>\n</ul>\n\n<h3>Who am I looking for?</h3>\n\n<ul>\n<li>Better score than public kernels</li>\n<li>Or previous competitions medals</li>\n<li>Ability to read R code </li>\n</ul>\n\n<p>Jacques,</p>",
      "rawMarkdown": "Hello everyone,\n\nI'm looking for a last minute teammate(s) for this competition. \n\n### Which language am I using?\n\nMy code base is in R for this competition. However I know how to code on Python too. See my Python solution for Instacart where I finished 52nd / top2 % \nhttps://github.com/jacquespeeters/instacart-market-basket-analysis\n\n###What is my workspace setup?\n\nI'm prototyping on my laptop when I'm training on hundreds thousands of lines. On millions lines I'm working on Google Cloud Platform. \n\n###What I have done on this competition?\n\nI had a close look at Ceshine kernels. However I reimplemented it enterely by myself in R with mostly dplyr style in order to be flexible and readable. Only focused on feature engineering, no model parameter tuning. However my solution isn't much better than public kernels, (0.53) and I'm quite disapointed. I'm wondering if I have a problem in my code which I don't spot. \n\n###What am I looking for? \n\n* Code review\n* Ideas sharing\n\n###Who am I looking for?\n\n* Better score than public kernels\n* Or previous competitions medals\n* Ability to read R code \n\nJacques,"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "264813": "Hello everyone,\n\nI'm looking for a last minute teammate(s) for this competition. \n\n### Which language am I using?\n\nMy code base is in R for this competition. However I know how to code on Python too. See my Python solution for Instacart where I finished 52nd / top2 % \nhttps://github.com/jacquespeeters/instacart-market-basket-analysis\n\n###What is my workspace setup?\n\nI'm prototyping on my laptop when I'm training on hundreds thousands of lines. On millions lines I'm working on Google Cloud Platform. \n\n###What I have done on this competition?\n\nI had a close look at Ceshine kernels. However I reimplemented it enterely by myself in R with mostly dplyr style in order to be flexible and readable. Only focused on feature engineering, no model parameter tuning. However my solution isn't much better than public kernels, (0.53) and I'm quite disapointed. I'm wondering if I have a problem in my code which I don't spot. \n\n###What am I looking for? \n\n* Code review\n* Ideas sharing\n\n###Who am I looking for?\n\n* Better score than public kernels\n* Or previous competitions medals\n* Ability to read R code \n\nJacques,"
  }
}