{
  "id": 99029,
  "title": "Even faster distance computation",
  "url": "/competitions/champs-scalar-coupling/discussion/99029",
  "author_name": "CPMP",
  "post_date": "2019-07-08T11:17:16.226000",
  "votes": 7,
  "comment_count": 0,
  "views": 0,
  "content": "<p>This is most probably not very useful anymore, but I'm sharing in case some find it useful still.</p>\n\n<p>In <a href=\"https://www.kaggle.com/seriousran/just-speed-up-calculate-distance-from-benchmark\">Just speed up calculate distance from Benchmark</a> kernel by <a href=\"https://www.kaggle.com/seriousran\">Chanran Kim</a>, <br> a faster way to calculate distance is provided. <br></p>\n\n<p>However, the computation requires data preprocessing.  If we include data preprocessing then there is a way to gain about 2x speedup again.</p>\n\n<p>Using Chanran Kim way:</p>\n\n<blockquote>\n  <p>CPU times: user 10 s, sys: 8.75 s, total: 18.8 s\n  Wall time: 18.8 s</p>\n</blockquote>\n\n<p>To:</p>\n\n<blockquote>\n  <p>CPU times: user 6.43 s, sys: 2.82 s, total: 9.25 s\n  Wall time: 9.14 s</p>\n</blockquote>\n\n<p>Kernel <a href=\"https://www.kaggle.com/cpmpml/even-faster-distance-computation?scriptVersionId=16895304\">here</a></p>",
  "messages": [
    {
      "id": 570489,
      "postDate": "2019-07-08T11:17:16.227Z",
      "content": "<p>This is most probably not very useful anymore, but I'm sharing in case some find it useful still.</p>\n\n<p>In <a href=\"https://www.kaggle.com/seriousran/just-speed-up-calculate-distance-from-benchmark\">Just speed up calculate distance from Benchmark</a> kernel by <a href=\"https://www.kaggle.com/seriousran\">Chanran Kim</a>, <br> a faster way to calculate distance is provided. <br></p>\n\n<p>However, the computation requires data preprocessing.  If we include data preprocessing then there is a way to gain about 2x speedup again.</p>\n\n<p>Using Chanran Kim way:</p>\n\n<blockquote>\n  <p>CPU times: user 10 s, sys: 8.75 s, total: 18.8 s\n  Wall time: 18.8 s</p>\n</blockquote>\n\n<p>To:</p>\n\n<blockquote>\n  <p>CPU times: user 6.43 s, sys: 2.82 s, total: 9.25 s\n  Wall time: 9.14 s</p>\n</blockquote>\n\n<p>Kernel <a href=\"https://www.kaggle.com/cpmpml/even-faster-distance-computation?scriptVersionId=16895304\">here</a></p>",
      "rawMarkdown": "This is most probably not very useful anymore, but I'm sharing in case some find it useful still.\n\nIn [Just speed up calculate distance from Benchmark](https://www.kaggle.com/seriousran/just-speed-up-calculate-distance-from-benchmark) kernel by [Chanran Kim](https://www.kaggle.com/seriousran), <br> a faster way to calculate distance is provided. <br>\n\nHowever, the computation requires data preprocessing.  If we include data preprocessing then there is a way to gain about 2x speedup again.\n\nUsing Chanran Kim way:\n&gt; CPU times: user 10 s, sys: 8.75 s, total: 18.8 s\n&gt; Wall time: 18.8 s\n\nTo:\n\n&gt; CPU times: user 6.43 s, sys: 2.82 s, total: 9.25 s\n&gt; Wall time: 9.14 s\n\nKernel [here](https://www.kaggle.com/cpmpml/even-faster-distance-computation?scriptVersionId=16895304)",
      "votes": 7
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "570489": "This is most probably not very useful anymore, but I'm sharing in case some find it useful still.\n\nIn [Just speed up calculate distance from Benchmark](https://www.kaggle.com/seriousran/just-speed-up-calculate-distance-from-benchmark) kernel by [Chanran Kim](https://www.kaggle.com/seriousran), <br> a faster way to calculate distance is provided. <br>\n\nHowever, the computation requires data preprocessing.  If we include data preprocessing then there is a way to gain about 2x speedup again.\n\nUsing Chanran Kim way:\n&gt; CPU times: user 10 s, sys: 8.75 s, total: 18.8 s\n&gt; Wall time: 18.8 s\n\nTo:\n\n&gt; CPU times: user 6.43 s, sys: 2.82 s, total: 9.25 s\n&gt; Wall time: 9.14 s\n\nKernel [here](https://www.kaggle.com/cpmpml/even-faster-distance-computation?scriptVersionId=16895304)"
  }
}