{
  "id": 52933,
  "title": "Pandas on ray",
  "url": "/competitions/talkingdata-adtracking-fraud-detection/discussion/52933",
  "author_name": "Pony",
  "post_date": "2018-03-25T02:31:59.524000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>Hey guys!  Have you ever heard about <a href=\"http://ray.readthedocs.io/en/latest/index.html\">ray</a>? It is \"a flexible, high-performance distributed execution framework\" accelerating calculation, which is designed by UC Berkeley and not yet finished. </p>\n\n<p>In its document, it says \"users can continue using their previous Pandas notebooks while experiencing a considerable speedup from Pandas on Ray, even on a single machine\". But it didn't not seem that easy for me. Many attributed of the old pandas are limited for me, which is not the same with other guys who posed \"how amazing ray is\". </p>\n\n<p>So, did I do something wrong when install it? I followed the official documents.</p>",
  "messages": [
    {
      "id": 302930,
      "postDate": "2018-03-25T02:31:59.523Z",
      "content": "<p>Hey guys!  Have you ever heard about <a href=\"http://ray.readthedocs.io/en/latest/index.html\">ray</a>? It is \"a flexible, high-performance distributed execution framework\" accelerating calculation, which is designed by UC Berkeley and not yet finished. </p>\n\n<p>In its document, it says \"users can continue using their previous Pandas notebooks while experiencing a considerable speedup from Pandas on Ray, even on a single machine\". But it didn't not seem that easy for me. Many attributed of the old pandas are limited for me, which is not the same with other guys who posed \"how amazing ray is\". </p>\n\n<p>So, did I do something wrong when install it? I followed the official documents.</p>",
      "rawMarkdown": "Hey guys!  Have you ever heard about [ray][1]? It is \"a flexible, high-performance distributed execution framework\" accelerating calculation, which is designed by UC Berkeley and not yet finished. \n\nIn its document, it says \"users can continue using their previous Pandas notebooks while experiencing a considerable speedup from Pandas on Ray, even on a single machine\". But it didn't not seem that easy for me. Many attributed of the old pandas are limited for me, which is not the same with other guys who posed \"how amazing ray is\". \n\nSo, did I do something wrong when install it? I followed the official documents.\n\n\n  [1]: http://ray.readthedocs.io/en/latest/index.html"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "302930": "Hey guys!  Have you ever heard about [ray][1]? It is \"a flexible, high-performance distributed execution framework\" accelerating calculation, which is designed by UC Berkeley and not yet finished. \n\nIn its document, it says \"users can continue using their previous Pandas notebooks while experiencing a considerable speedup from Pandas on Ray, even on a single machine\". But it didn't not seem that easy for me. Many attributed of the old pandas are limited for me, which is not the same with other guys who posed \"how amazing ray is\". \n\nSo, did I do something wrong when install it? I followed the official documents.\n\n\n  [1]: http://ray.readthedocs.io/en/latest/index.html"
  }
}