{
  "id": 122263,
  "title": "There's equal train data for some meters & some building_id's",
  "url": "/competitions/ashrae-energy-prediction/discussion/122263",
  "author_name": "Artyom Vorobyov",
  "post_date": "2019-12-18T22:49:44.502000",
  "votes": 10,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Interestingly I've found, that there are similar meter measurements for entire train period for some <code>building_id</code>s. The list is below. The measurements are equal, they are just scaled by the <code>square_feet</code>. For some matching paris <code>primary_use</code>s are different.\n<code>\nm0 equals: [1225, 1226]\nm0 equals: [1233, 1234]\nm0 equals: [1277, 1278, 1279]\nm0 equals: [1296, 1297, 1299]\nm1 equals: [1223, 1229] - partially matches\nm1 equals: [1291, 1293, 1300]\nm1 equals: [1294, 1302]\nm2 equals: [1225, 1226]\nm2 equals: [1294, 1295]\nm2 equals: [1296, 1297, 1298, 1299]\nm3 equals: [1223, 1229, 1231] --- 1223/1231 has a segment filled with 0s, while 1229 hasn't\nm3 equals: [1241, 1253] --- 1241 has a segment filled with 0s, while 1253 hasn't\nm3 equals: [1272, 1273] --- 1272 has a segment filled with 0s, while 1273 hasn't\nm3 equals: [1275, 1276]\nm3 equals: [1294, 1295, 1296, 1297, 1300]\nm3 equals: [1318, 1319]\n</code></p>",
  "messages": [
    {
      "id": 698184,
      "postDate": "2019-12-18T22:49:44.503Z",
      "content": "<p>Interestingly I've found, that there are similar meter measurements for entire train period for some <code>building_id</code>s. The list is below. The measurements are equal, they are just scaled by the <code>square_feet</code>. For some matching paris <code>primary_use</code>s are different.\n<code>\nm0 equals: [1225, 1226]\nm0 equals: [1233, 1234]\nm0 equals: [1277, 1278, 1279]\nm0 equals: [1296, 1297, 1299]\nm1 equals: [1223, 1229] - partially matches\nm1 equals: [1291, 1293, 1300]\nm1 equals: [1294, 1302]\nm2 equals: [1225, 1226]\nm2 equals: [1294, 1295]\nm2 equals: [1296, 1297, 1298, 1299]\nm3 equals: [1223, 1229, 1231] --- 1223/1231 has a segment filled with 0s, while 1229 hasn't\nm3 equals: [1241, 1253] --- 1241 has a segment filled with 0s, while 1253 hasn't\nm3 equals: [1272, 1273] --- 1272 has a segment filled with 0s, while 1273 hasn't\nm3 equals: [1275, 1276]\nm3 equals: [1294, 1295, 1296, 1297, 1300]\nm3 equals: [1318, 1319]\n</code></p>",
      "rawMarkdown": "Interestingly I've found, that there are similar meter measurements for entire train period for some `building_id`s. The list is below. The measurements are equal, they are just scaled by the `square_feet`. For some matching paris `primary_use`s are different.\n```\nm0 equals: [1225, 1226]\nm0 equals: [1233, 1234]\nm0 equals: [1277, 1278, 1279]\nm0 equals: [1296, 1297, 1299]\nm1 equals: [1223, 1229] - partially matches\nm1 equals: [1291, 1293, 1300]\nm1 equals: [1294, 1302]\nm2 equals: [1225, 1226]\nm2 equals: [1294, 1295]\nm2 equals: [1296, 1297, 1298, 1299]\nm3 equals: [1223, 1229, 1231] --- 1223/1231 has a segment filled with 0s, while 1229 hasn't\nm3 equals: [1241, 1253] --- 1241 has a segment filled with 0s, while 1253 hasn't\nm3 equals: [1272, 1273] --- 1272 has a segment filled with 0s, while 1273 hasn't\nm3 equals: [1275, 1276]\nm3 equals: [1294, 1295, 1296, 1297, 1300]\nm3 equals: [1318, 1319]\n```",
      "votes": 9
    },
    {
      "id": 698331,
      "postDate": "2019-12-19T04:45:02.283Z",
      "content": "<p>There's one more notice. Meter 3, <code>building_id</code> 1241 is filled not with zeros. It has left part the same as <code>building_id</code> 1253, but the right part of 1241 is filled with very low numbers. Seems like they're taken from the other <code>building_id</code>.</p>",
      "rawMarkdown": "There's one more notice. Meter 3, `building_id` 1241 is filled not with zeros. It has left part the same as `building_id` 1253, but the right part of 1241 is filled with very low numbers. Seems like they're taken from the other `building_id`."
    },
    {
      "id": 698196,
      "postDate": "2019-12-18T23:07:14.687Z",
      "content": "<p>really nice finding! but it is too late for me to use.</p>",
      "rawMarkdown": "really nice finding! but it is too late for me to use."
    },
    {
      "id": 698758,
      "postDate": "2019-12-19T17:01:04.530Z",
      "rawMarkdown": "",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 698331,
      "author_name": "Artyom Vorobyov",
      "author_url": "",
      "post_date": "2019-12-19T04:45:02.283000",
      "content": "<p>There's one more notice. Meter 3, <code>building_id</code> 1241 is filled not with zeros. It has left part the same as <code>building_id</code> 1253, but the right part of 1241 is filled with very low numbers. Seems like they're taken from the other <code>building_id</code>.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 698196,
      "author_name": "eagle4",
      "author_url": "",
      "post_date": "2019-12-18T23:07:14.687000",
      "content": "<p>really nice finding! but it is too late for me to use.</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 698758,
      "author_name": "",
      "author_url": "",
      "post_date": "2019-12-19T17:01:04.530000",
      "content": "",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "698184": "Interestingly I've found, that there are similar meter measurements for entire train period for some `building_id`s. The list is below. The measurements are equal, they are just scaled by the `square_feet`. For some matching paris `primary_use`s are different.\n```\nm0 equals: [1225, 1226]\nm0 equals: [1233, 1234]\nm0 equals: [1277, 1278, 1279]\nm0 equals: [1296, 1297, 1299]\nm1 equals: [1223, 1229] - partially matches\nm1 equals: [1291, 1293, 1300]\nm1 equals: [1294, 1302]\nm2 equals: [1225, 1226]\nm2 equals: [1294, 1295]\nm2 equals: [1296, 1297, 1298, 1299]\nm3 equals: [1223, 1229, 1231] --- 1223/1231 has a segment filled with 0s, while 1229 hasn't\nm3 equals: [1241, 1253] --- 1241 has a segment filled with 0s, while 1253 hasn't\nm3 equals: [1272, 1273] --- 1272 has a segment filled with 0s, while 1273 hasn't\nm3 equals: [1275, 1276]\nm3 equals: [1294, 1295, 1296, 1297, 1300]\nm3 equals: [1318, 1319]\n```",
    "698331": "There's one more notice. Meter 3, `building_id` 1241 is filled not with zeros. It has left part the same as `building_id` 1253, but the right part of 1241 is filled with very low numbers. Seems like they're taken from the other `building_id`.",
    "698196": "really nice finding! but it is too late for me to use.",
    "698758": ""
  }
}