{
  "id": 6548,
  "title": "R appears to have a pretty bad leak with the hash lib",
  "url": "/competitions/facebook-recruiting-iii-keyword-extraction/discussion/6548",
  "author_name": "David Cho",
  "post_date": "2013-12-06T06:29:42.047000",
  "votes": 0,
  "comment_count": 0,
  "views": 800,
  "content": "<p>I know, I know...why am I using R for a large dataset like this? &nbsp;I already went too far down this road to changeover, besides it's not too bad, there are plenty of R libraries, but the hash library is really slowing me down, which is counter-intuitive. &nbsp;The hash lookups are fast but over time, it slows down and I think its from a memory leak within R or the hash library or both. &nbsp;Is anyone else seeing this? I usually see a growing memory footprint over the same sized hash, causing performance slowdown. I have to restart R and continue from where I left off....hash tables should NOT have memory leaks...R is not that mature I guess.&nbsp;</p>",
  "messages": [
    {
      "id": 35898,
      "postDate": "2013-12-06T06:29:42.047Z",
      "content": "<p>I know, I know...why am I using R for a large dataset like this? &nbsp;I already went too far down this road to changeover, besides it's not too bad, there are plenty of R libraries, but the hash library is really slowing me down, which is counter-intuitive. &nbsp;The hash lookups are fast but over time, it slows down and I think its from a memory leak within R or the hash library or both. &nbsp;Is anyone else seeing this? I usually see a growing memory footprint over the same sized hash, causing performance slowdown. I have to restart R and continue from where I left off....hash tables should NOT have memory leaks...R is not that mature I guess.&nbsp;</p>"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "35898": ""
  }
}