{
  "id": 391749,
  "title": "Performance Questions",
  "url": "/competitions/asl-signs/discussion/391749",
  "author_name": "",
  "post_date": "2023-03-02T15:02:15.501595100Z",
  "votes": null,
  "comment_count": 4,
  "views": 0,
  "content": "<p>\"With less than 100 milliseconds of latency per video\" </p>\n<p>\"We allow an additional 10 minute buffer for loading the data and miscellaneous overhead.\"</p>\n<p>\"Expect to see approximately 40,000 videos in the test set.\"</p>\n<p>So does this mean we are just given ~4600 seconds to complete everything or is the latency a hard maxima per sample? </p>\n<p>Calc:<br>\n ~40 000 * 100ms + 10 min * 60 sec/min * 1 000 ms/s = 4 600 000 ms = 4600 seconds </p>\n<p>Thanks!</p>\n<p>Feel free to add any other perf questions to the thread </p>",
  "messages": [
    {
      "id": "2165974",
      "postDate": "03/02/2023 15:02:15",
      "content": "<p>\"With less than 100 milliseconds of latency per video\" </p>\n<p>\"We allow an additional 10 minute buffer for loading the data and miscellaneous overhead.\"</p>\n<p>\"Expect to see approximately 40,000 videos in the test set.\"</p>\n<p>So does this mean we are just given ~4600 seconds to complete everything or is the latency a hard maxima per sample? </p>\n<p>Calc:<br>\n ~40 000 * 100ms + 10 min * 60 sec/min * 1 000 ms/s = 4 600 000 ms = 4600 seconds </p>\n<p>Thanks!</p>\n<p>Feel free to add any other perf questions to the thread </p>",
      "rawMarkdown": "\"With less than 100 milliseconds of latency per video\" \n\n\"We allow an additional 10 minute buffer for loading the data and miscellaneous overhead.\"\n\n\"Expect to see approximately 40,000 videos in the test set.\"\n\nSo does this mean we are just given ~4600 seconds to complete everything or is the latency a hard maxima per sample? \n\nCalc:\n ~40 000 * 100ms + 10 min * 60 sec/min * 1 000 ms/s = 4 600 000 ms = 4600 seconds \n\nThanks!\n\nFeel free to add any other perf questions to the thread",
      "votes": null
    },
    {
      "id": "2166915",
      "postDate": "03/03/2023 05:43:39",
      "content": "<p>\"inference with less than 100 milliseconds of latency per video\"</p>\n<p>is this average timming?<br>\nthe longest video could be &gt;500 frames?</p>",
      "rawMarkdown": "\"inference with less than 100 milliseconds of latency per video\"\n\nis this average timming?\nthe longest video could be >500 frames?",
      "votes": null
    },
    {
      "id": "2169790",
      "postDate": "03/05/2023 12:41:46",
      "content": "<p>I would imagine it's 100 milliseconds average as otherwise the mention of 40.000 plus the extra 10 minutes makes less sense.</p>",
      "rawMarkdown": "I would imagine it's 100 milliseconds average as otherwise the mention of 40.000 plus the extra 10 minutes makes less sense.",
      "votes": null
    },
    {
      "id": "2171653",
      "postDate": "03/07/2023 01:02:07",
      "content": "<p>It's on average. I've updated the evaluation page to clarify that point.</p>",
      "rawMarkdown": "It's on average. I've updated the evaluation page to clarify that point.",
      "votes": null
    },
    {
      "id": "2172446",
      "postDate": "03/07/2023 14:37:40",
      "content": "<p>Thanks! I'm new to Kaggle so forgive the possibly obvious question</p>",
      "rawMarkdown": "Thanks! I'm new to Kaggle so forgive the possibly obvious question",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2166915,
      "author_name": "hengck23",
      "author_url": "",
      "post_date": "03/03/2023 05:43:39",
      "content": "<p>\"inference with less than 100 milliseconds of latency per video\"</p>\n<p>is this average timming?<br>\nthe longest video could be &gt;500 frames?</p>",
      "votes": null,
      "replies": [
        {
          "id": 2169790,
          "author_name": "tom3141592",
          "author_url": "",
          "post_date": "03/05/2023 12:41:46",
          "content": "<p>I would imagine it's 100 milliseconds average as otherwise the mention of 40.000 plus the extra 10 minutes makes less sense.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 2171653,
          "author_name": "sohier",
          "author_url": "",
          "post_date": "03/07/2023 01:02:07",
          "content": "<p>It's on average. I've updated the evaluation page to clarify that point.</p>",
          "votes": null,
          "replies": [
            {
              "id": 2172446,
              "author_name": "chrishunter3",
              "author_url": "",
              "post_date": "03/07/2023 14:37:40",
              "content": "<p>Thanks! I'm new to Kaggle so forgive the possibly obvious question</p>",
              "votes": null,
              "replies": []
            }
          ]
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2165974": "\"With less than 100 milliseconds of latency per video\" \n\n\"We allow an additional 10 minute buffer for loading the data and miscellaneous overhead.\"\n\n\"Expect to see approximately 40,000 videos in the test set.\"\n\nSo does this mean we are just given ~4600 seconds to complete everything or is the latency a hard maxima per sample? \n\nCalc:\n ~40 000 * 100ms + 10 min * 60 sec/min * 1 000 ms/s = 4 600 000 ms = 4600 seconds \n\nThanks!\n\nFeel free to add any other perf questions to the thread",
    "2166915": "\"inference with less than 100 milliseconds of latency per video\"\n\nis this average timming?\nthe longest video could be >500 frames?",
    "2169790": "I would imagine it's 100 milliseconds average as otherwise the mention of 40.000 plus the extra 10 minutes makes less sense.",
    "2171653": "It's on average. I've updated the evaluation page to clarify that point.",
    "2172446": "Thanks! I'm new to Kaggle so forgive the possibly obvious question"
  },
  "source": "meta"
}