{
  "id": 100242,
  "title": "Ultra Fast Distance Matrix Computation",
  "url": "/competitions/champs-scalar-coupling/discussion/100242",
  "author_name": "",
  "post_date": "2019-07-17T10:58:29.793008600Z",
  "votes": 29,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Distance matrix between atoms of a single molecule are used in many public kernels to compute features, for instance Coulomb interaction, Van de Walls interaction, and Yukawa interactions, </p>\n\n<p>The already fast version used in <a href=\"https://www.kaggle.com/rio114/coulomb-interaction-speed-up\">coulomb_interaction - speed up!</a> takes 2 minutes for all molecules.  </p>\n\n<p><a href=\"https://www.kaggle.com/cpmpml/ultra-fast-distance-matrix-computation\">here</a> I provide a code that runs in 0.4 seconds, i.e. 250 times faster.</p>\n\n<p>This speedup is nice.  Moreover, the code optimization technique used in this kernel is rather generic and can be reused in other context.</p>\n\n<p>Updated with better results.</p>",
  "messages": [
    {
      "id": "578109",
      "postDate": "07/17/2019 10:58:29",
      "content": "<p>Distance matrix between atoms of a single molecule are used in many public kernels to compute features, for instance Coulomb interaction, Van de Walls interaction, and Yukawa interactions, </p>\n\n<p>The already fast version used in <a href=\"https://www.kaggle.com/rio114/coulomb-interaction-speed-up\">coulomb_interaction - speed up!</a> takes 2 minutes for all molecules.  </p>\n\n<p><a href=\"https://www.kaggle.com/cpmpml/ultra-fast-distance-matrix-computation\">here</a> I provide a code that runs in 0.4 seconds, i.e. 250 times faster.</p>\n\n<p>This speedup is nice.  Moreover, the code optimization technique used in this kernel is rather generic and can be reused in other context.</p>\n\n<p>Updated with better results.</p>",
      "rawMarkdown": "Distance matrix between atoms of a single molecule are used in many public kernels to compute features, for instance Coulomb interaction, Van de Walls interaction, and Yukawa interactions, \n\nThe already fast version used in [coulomb_interaction - speed up!](https://www.kaggle.com/rio114/coulomb-interaction-speed-up) takes 2 minutes for all molecules.  \n\n [here](https://www.kaggle.com/cpmpml/ultra-fast-distance-matrix-computation) I provide a code that runs in 0.4 seconds, i.e. 250 times faster.\n\nThis speedup is nice.  Moreover, the code optimization technique used in this kernel is rather generic and can be reused in other context.\n\nUpdated with better results.",
      "votes": null
    },
    {
      "id": "578430",
      "postDate": "07/17/2019 17:19:38",
      "content": "<p>As usual, a top level contribution by you CPMP, thanks for sharing!  I believe a good computational speed is indeed required for this competition :-)</p>",
      "rawMarkdown": "As usual, a top level contribution by you CPMP, thanks for sharing!  I believe a good computational speed is indeed required for this competition :-)",
      "votes": null
    },
    {
      "id": "578436",
      "postDate": "07/17/2019 17:29:35",
      "content": "<p>Thanks!</p>",
      "rawMarkdown": "Thanks!",
      "votes": null
    },
    {
      "id": "578793",
      "postDate": "07/18/2019 06:51:07",
      "content": "<p>Updated the notebook with a code that runs in less than 0.4 seconds, which is a 250x speedup compared to the original code.</p>",
      "rawMarkdown": "Updated the notebook with a code that runs in less than 0.4 seconds, which is a 250x speedup compared to the original code.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 578430,
      "author_name": "abrial",
      "author_url": "",
      "post_date": "07/17/2019 17:19:38",
      "content": "<p>As usual, a top level contribution by you CPMP, thanks for sharing!  I believe a good computational speed is indeed required for this competition :-)</p>",
      "votes": null,
      "replies": [
        {
          "id": 578436,
          "author_name": "cpmpml",
          "author_url": "",
          "post_date": "07/17/2019 17:29:35",
          "content": "<p>Thanks!</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 578793,
      "author_name": "cpmpml",
      "author_url": "",
      "post_date": "07/18/2019 06:51:07",
      "content": "<p>Updated the notebook with a code that runs in less than 0.4 seconds, which is a 250x speedup compared to the original code.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "578109": "Distance matrix between atoms of a single molecule are used in many public kernels to compute features, for instance Coulomb interaction, Van de Walls interaction, and Yukawa interactions, \n\nThe already fast version used in [coulomb_interaction - speed up!](https://www.kaggle.com/rio114/coulomb-interaction-speed-up) takes 2 minutes for all molecules.  \n\n [here](https://www.kaggle.com/cpmpml/ultra-fast-distance-matrix-computation) I provide a code that runs in 0.4 seconds, i.e. 250 times faster.\n\nThis speedup is nice.  Moreover, the code optimization technique used in this kernel is rather generic and can be reused in other context.\n\nUpdated with better results.",
    "578430": "As usual, a top level contribution by you CPMP, thanks for sharing!  I believe a good computational speed is indeed required for this competition :-)",
    "578436": "Thanks!",
    "578793": "Updated the notebook with a code that runs in less than 0.4 seconds, which is a 250x speedup compared to the original code."
  },
  "source": "meta"
}