{
  "id": 6136,
  "title": "Two appliances on in the backend solution for the same signal in the data",
  "url": "/competitions/belkin-energy-disaggregation-competition/discussion/6136",
  "author_name": "",
  "post_date": "2013-10-28T11:33:11.293Z",
  "votes": null,
  "comment_count": 2,
  "views": 1616,
  "content": "<p>Has anyone else seen this?</p>\n<p>I tried two similar appliances in different submissions to see which one was scored as right. &nbsp;Now that I am sorting things out and have more confidence about the other appliances in those submissions, it looks like both appliances are scored as &quot;On&quot; in the backend solution even though the size of the signal fits one but not both.</p>\n<p>Oh, well ...</p>",
  "messages": [
    {
      "id": "32770",
      "postDate": "10/28/2013 11:33:11",
      "content": "<p>Has anyone else seen this?</p>\n<p>I tried two similar appliances in different submissions to see which one was scored as right. &nbsp;Now that I am sorting things out and have more confidence about the other appliances in those submissions, it looks like both appliances are scored as &quot;On&quot; in the backend solution even though the size of the signal fits one but not both.</p>\n<p>Oh, well ...</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "32943",
      "postDate": "10/31/2013 15:31:32",
      "content": "<p>Hi Noam,</p>\n<p>&nbsp;</p>\n<p>I think I know which appliances you're talking about. I&nbsp;wish I could have replied earlier, but you know, we were still competing against each other.</p>\n<p>I think you're talking about appliances 17 and 19 of House 1. One is GR PS4 and the other is GR LCD TV. If that is true, this is what I saw. It happens that most of the time both are turned on and turned off at the same time, so both score as on.&nbsp;</p>\n<p>&nbsp;</p>\n<p>In the private leaderboard, those two appliances are on in three intervals of 393, 188, and 77 minutes, which in total is 2*658 minutes, which makes a significant difference.&nbsp;Those ones may be some that you missed.</p>\n<p>[quote=Noam Tene;32923]</p>\n<p>I am wondering if there is an easy way to figure out which appliances you found that I either missed or chose not to mark because I was not confident about them.</p>\n<p>[/quote]</p>\n<p>&nbsp;</p>\n<p>&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    },
    {
      "id": "32961",
      "postDate": "10/31/2013 18:04:58",
      "content": "<p>Luis,&nbsp;</p>\n<p>It's funny that you mentioned that.</p>\n<p>You just answered two questions that I asked other threads in response to a question that I did not think was related to either. &nbsp;</p>\n<p>In this thread I was actually referring to the two H1 vaccum cleaners (appliances 8 and 32). &nbsp; Each vaccum cleaner is tagged only once in the tagged data and I believe these two tags were not well marked and are very confusing. &nbsp;The only thing I could figure out from them was that both Vaccum cleaners had a very strong and somewhat similar HF signature that looked very different from other HF signatures.</p>\n<p>There are 4 instances where those HF signatures appear in the public fold and my submissions initially showed that in the back end, the &quot;Portable Vaccum&quot; (appliance 32)&quot; was marked as on for at least three of these time periods.</p>\n<p>As time went on I was surprised to find out that the fourth occurrence which lasts only 2 minutes had both Vaccums marked as &quot;On&quot; in the backend solution and this is what I was referring to when I posted this thread.</p>\n<p>&nbsp;</p>",
      "rawMarkdown": "",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 32943,
      "author_name": "luistp001",
      "author_url": "",
      "post_date": "10/31/2013 15:31:32",
      "content": "<p>Hi Noam,</p>\n<p>&nbsp;</p>\n<p>I think I know which appliances you're talking about. I&nbsp;wish I could have replied earlier, but you know, we were still competing against each other.</p>\n<p>I think you're talking about appliances 17 and 19 of House 1. One is GR PS4 and the other is GR LCD TV. If that is true, this is what I saw. It happens that most of the time both are turned on and turned off at the same time, so both score as on.&nbsp;</p>\n<p>&nbsp;</p>\n<p>In the private leaderboard, those two appliances are on in three intervals of 393, 188, and 77 minutes, which in total is 2*658 minutes, which makes a significant difference.&nbsp;Those ones may be some that you missed.</p>\n<p>[quote=Noam Tene;32923]</p>\n<p>I am wondering if there is an easy way to figure out which appliances you found that I either missed or chose not to mark because I was not confident about them.</p>\n<p>[/quote]</p>\n<p>&nbsp;</p>\n<p>&nbsp;</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 32961,
      "author_name": "noamtene",
      "author_url": "",
      "post_date": "10/31/2013 18:04:58",
      "content": "<p>Luis,&nbsp;</p>\n<p>It's funny that you mentioned that.</p>\n<p>You just answered two questions that I asked other threads in response to a question that I did not think was related to either. &nbsp;</p>\n<p>In this thread I was actually referring to the two H1 vaccum cleaners (appliances 8 and 32). &nbsp; Each vaccum cleaner is tagged only once in the tagged data and I believe these two tags were not well marked and are very confusing. &nbsp;The only thing I could figure out from them was that both Vaccum cleaners had a very strong and somewhat similar HF signature that looked very different from other HF signatures.</p>\n<p>There are 4 instances where those HF signatures appear in the public fold and my submissions initially showed that in the back end, the &quot;Portable Vaccum&quot; (appliance 32)&quot; was marked as on for at least three of these time periods.</p>\n<p>As time went on I was surprised to find out that the fourth occurrence which lasts only 2 minutes had both Vaccums marked as &quot;On&quot; in the backend solution and this is what I was referring to when I posted this thread.</p>\n<p>&nbsp;</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "32770": "",
    "32943": "",
    "32961": ""
  },
  "source": "meta"
}