{
  "id": 420457,
  "title": "Any idea about how to use NMS on Yolov8?",
  "url": "/competitions/hubmap-hacking-the-human-vasculature/discussion/420457",
  "author_name": "",
  "post_date": "2023-06-30T22:45:58.562230400Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi, does anybody know about how to use NMS on Yolov8?</p>\n<p>Thanks in advance!</p>",
  "messages": [
    {
      "id": "2324824",
      "postDate": "06/30/2023 22:45:58",
      "content": "<p>Hi, does anybody know about how to use NMS on Yolov8?</p>\n<p>Thanks in advance!</p>",
      "rawMarkdown": "Hi, does anybody know about how to use NMS on Yolov8?\n\nThanks in advance!",
      "votes": null
    },
    {
      "id": "2326702",
      "postDate": "07/02/2023 10:30:11",
      "content": "<p>res = preds[0]<br>\nboxes = res.boxes.data<br>\nmasks = res.masks.data<br>\np = ops.non_max_suppression(boxes,<br>\n                            classes=0)</p>\n<p>And the argument will be :</p>\n<p>def non_max_suppression(<br>\n        prediction,<br>\n        conf_thres=0.25,<br>\n        iou_thres=0.45,<br>\n        classes=None,<br>\n        agnostic=False,<br>\n        multi_label=False,<br>\n        labels=(),<br>\n        max_det=300,<br>\n        nc=0,  # number of classes (optional)<br>\n        max_time_img=0.05,<br>\n        max_nms=30000,<br>\n        max_wh=7680,<br>\n)</p>",
      "rawMarkdown": "res = preds[0]\nboxes = res.boxes.data\nmasks = res.masks.data\np = ops.non_max_suppression(boxes,\n                            classes=0)\n\nAnd the argument will be :\n\ndef non_max_suppression(\n        prediction,\n        conf_thres=0.25,\n        iou_thres=0.45,\n        classes=None,\n        agnostic=False,\n        multi_label=False,\n        labels=(),\n        max_det=300,\n        nc=0,  # number of classes (optional)\n        max_time_img=0.05,\n        max_nms=30000,\n        max_wh=7680,\n)",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2326702,
      "author_name": "ahmedabdelhamied2501",
      "author_url": "",
      "post_date": "07/02/2023 10:30:11",
      "content": "<p>res = preds[0]<br>\nboxes = res.boxes.data<br>\nmasks = res.masks.data<br>\np = ops.non_max_suppression(boxes,<br>\n                            classes=0)</p>\n<p>And the argument will be :</p>\n<p>def non_max_suppression(<br>\n        prediction,<br>\n        conf_thres=0.25,<br>\n        iou_thres=0.45,<br>\n        classes=None,<br>\n        agnostic=False,<br>\n        multi_label=False,<br>\n        labels=(),<br>\n        max_det=300,<br>\n        nc=0,  # number of classes (optional)<br>\n        max_time_img=0.05,<br>\n        max_nms=30000,<br>\n        max_wh=7680,<br>\n)</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2324824": "Hi, does anybody know about how to use NMS on Yolov8?\n\nThanks in advance!",
    "2326702": "res = preds[0]\nboxes = res.boxes.data\nmasks = res.masks.data\np = ops.non_max_suppression(boxes,\n                            classes=0)\n\nAnd the argument will be :\n\ndef non_max_suppression(\n        prediction,\n        conf_thres=0.25,\n        iou_thres=0.45,\n        classes=None,\n        agnostic=False,\n        multi_label=False,\n        labels=(),\n        max_det=300,\n        nc=0,  # number of classes (optional)\n        max_time_img=0.05,\n        max_nms=30000,\n        max_wh=7680,\n)"
  },
  "source": "meta"
}