{
  "id": 203498,
  "title": "Is there anyone using FPs?",
  "url": "/competitions/rfcx-species-audio-detection/discussion/203498",
  "author_name": "",
  "post_date": "2020-12-15T14:34:39.218882800Z",
  "votes": 9,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hi everyone,</p>\n<p>have you seen any improvement when using FPs information during training or test? I'm wondering if they are too dissimilar from the TPs in order to use a contrastive learning approach.</p>\n<p>Best,</p>\n<p>Guglielmo</p>",
  "messages": [
    {
      "id": "1113539",
      "postDate": "12/15/2020 14:34:39",
      "content": "<p>Hi everyone,</p>\n<p>have you seen any improvement when using FPs information during training or test? I'm wondering if they are too dissimilar from the TPs in order to use a contrastive learning approach.</p>\n<p>Best,</p>\n<p>Guglielmo</p>",
      "rawMarkdown": "Hi everyone,\n\nhave you seen any improvement when using FPs information during training or test? I'm wondering if they are too dissimilar from the TPs in order to use a contrastive learning approach.\n\nBest,\n\nGuglielmo",
      "votes": null
    },
    {
      "id": "1115478",
      "postDate": "12/16/2020 10:14:04",
      "content": "<p>I haven't seen improvements. But I will get back to it, there must be some juice to squeeze. <br>\nContrastive learning is computationally very intensive so I will (likely) skip it.</p>",
      "rawMarkdown": "I haven't seen improvements. But I will get back to it, there must be some juice to squeeze. \nContrastive learning is computationally very intensive so I will (likely) skip it.",
      "votes": null
    },
    {
      "id": "1116378",
      "postDate": "12/17/2020 05:56:44",
      "content": "<p>I have the same question. False positive just means that the species is not in the identified region, right?</p>",
      "rawMarkdown": "I have the same question. False positive just means that the species is not in the identified region, right?",
      "votes": null
    },
    {
      "id": "1116461",
      "postDate": "12/17/2020 07:31:48",
      "content": "<p>No, I guess FPs are positive labeled samples (labeled from an automatic algorithm), that are checked and rejected by human experts.</p>\n<p>Best,</p>\n<p>Guglielmo</p>",
      "rawMarkdown": "No, I guess FPs are positive labeled samples (labeled from an automatic algorithm), that are checked and rejected by human experts.\n\nBest,\n\nGuglielmo",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1115478,
      "author_name": "valanm",
      "author_url": "",
      "post_date": "12/16/2020 10:14:04",
      "content": "<p>I haven't seen improvements. But I will get back to it, there must be some juice to squeeze. <br>\nContrastive learning is computationally very intensive so I will (likely) skip it.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1116378,
      "author_name": "hazarre",
      "author_url": "",
      "post_date": "12/17/2020 05:56:44",
      "content": "<p>I have the same question. False positive just means that the species is not in the identified region, right?</p>",
      "votes": null,
      "replies": [
        {
          "id": 1116461,
          "author_name": "guglielmocamporese",
          "author_url": "",
          "post_date": "12/17/2020 07:31:48",
          "content": "<p>No, I guess FPs are positive labeled samples (labeled from an automatic algorithm), that are checked and rejected by human experts.</p>\n<p>Best,</p>\n<p>Guglielmo</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1113539": "Hi everyone,\n\nhave you seen any improvement when using FPs information during training or test? I'm wondering if they are too dissimilar from the TPs in order to use a contrastive learning approach.\n\nBest,\n\nGuglielmo",
    "1115478": "I haven't seen improvements. But I will get back to it, there must be some juice to squeeze. \nContrastive learning is computationally very intensive so I will (likely) skip it.",
    "1116378": "I have the same question. False positive just means that the species is not in the identified region, right?",
    "1116461": "No, I guess FPs are positive labeled samples (labeled from an automatic algorithm), that are checked and rejected by human experts.\n\nBest,\n\nGuglielmo"
  },
  "source": "meta"
}