{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Crowdpose example"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<https://github.com/jin-s13/xtcocoapi/blob/master/demos/demo_crowdpose.py>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "faster_coco_eval.__version__='1.6.4'\n"
     ]
    }
   ],
   "source": [
    "import logging\n",
    "import numpy as np\n",
    "import faster_coco_eval\n",
    "from faster_coco_eval import COCO, COCOeval_faster\n",
    "\n",
    "print(f\"{faster_coco_eval.__version__=}\")\n",
    "\n",
    "logging.root.setLevel(\"INFO\")\n",
    "logging.debug(\"Запись.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "gt_file = '../tests/dataset/example_crowdpose_val.json'\n",
    "preds = '../tests/dataset/example_crowdpose_preds.json'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "sigmas = np.array([\n",
    "            .79, .79, .72, .72, .62, .62, 1.07, 1.07, .87, .87, .89, .89, .79,\n",
    "            .79\n",
    "        ]) / 10.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:faster_coco_eval.core.cocoeval:Evaluate annotation type *keypoints_crowd*\n",
      "INFO:faster_coco_eval.core.cocoeval:COCOeval_opt.evaluate() finished...\n",
      "INFO:faster_coco_eval.core.cocoeval:DONE (t=0.00s).\n",
      "INFO:faster_coco_eval.core.cocoeval:Accumulating evaluation results...\n",
      "INFO:faster_coco_eval.core.cocoeval:COCOeval_opt.accumulate() finished...\n",
      "INFO:faster_coco_eval.core.cocoeval:DONE (t=0.00s).\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets= 20 ] = 0.788\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets= 20 ] = 0.988\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets= 20 ] = 0.731\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 20 ] = 0.822\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Recall     (AR) @[ IoU=0.50      | area=   all | maxDets= 20 ] = 1.000\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Recall     (AR) @[ IoU=0.75      | area=   all | maxDets= 20 ] = 0.778\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Precision  (AP) @[ IoU=0.50:0.95 | type=  easy | maxDets= 20 ] = 1.000\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Precision  (AP) @[ IoU=0.50:0.95 | type=medium | maxDets= 20 ] = 0.980\n",
      "INFO:faster_coco_eval.core.cocoeval: Average Precision  (AP) @[ IoU=0.50:0.95 | type=  hard | maxDets= 20 ] = 0.412\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "{'AP_all': 0.7877215935879303,\n",
       " 'AP_50': 0.9881188118811886,\n",
       " 'AP_75': 0.7314356435643564,\n",
       " 'AR_all': 0.8222222222222223,\n",
       " 'AR_50': 1.0,\n",
       " 'AR_75': 0.7777777777777778,\n",
       " 'AP_easy': 1.0,\n",
       " 'AP_medium': 0.9802,\n",
       " 'AP_hard': 0.4116}"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cocoGt = COCO(gt_file)\n",
    "cocoDt = cocoGt.loadRes(preds)\n",
    "cocoEval = COCOeval_faster(cocoGt, cocoDt, 'keypoints_crowd', kpt_oks_sigmas=sigmas, use_area=False)\n",
    "cocoEval.evaluate()\n",
    "cocoEval.accumulate()\n",
    "cocoEval.summarize()\n",
    "\n",
    "cocoEval.stats_as_dict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Orig Code\n",
    "\n",
    "```py\n",
    "from xtcocotools.coco import COCO\n",
    "from xtcocotools.cocoeval import COCOeval\n",
    "\n",
    "\n",
    "cocoGt = COCO(gt_file)\n",
    "cocoDt = cocoGt.loadRes(preds)\n",
    "cocoEval = COCOeval(cocoGt, cocoDt, 'keypoints_crowd', sigmas, use_area=False)\n",
    "cocoEval.evaluate()\n",
    "cocoEval.accumulate()\n",
    "cocoEval.summarize()\n",
    "```\n",
    "\n",
    "## Orig result\n",
    "\n",
    "```text\n",
    "loading annotations into memory...\n",
    "Done (t=0.00s)\n",
    "creating index...\n",
    "index created!\n",
    "Loading and preparing results...\n",
    "DONE (t=0.00s)\n",
    "creating index...\n",
    "index created!\n",
    "Running per image evaluation...\n",
    "Evaluate annotation type *keypoints_crowd*\n",
    "DONE (t=0.00s).\n",
    "Accumulating evaluation results...\n",
    "DONE (t=0.00s).\n",
    " Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets= 20 ] =  0.788\n",
    " Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets= 20 ] =  0.988\n",
    " Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets= 20 ] =  0.731\n",
    " Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 20 ] =  0.822\n",
    " Average Recall     (AR) @[ IoU=0.50      | area=   all | maxDets= 20 ] =  1.000\n",
    " Average Recall     (AR) @[ IoU=0.75      | area=   all | maxDets= 20 ] =  0.778\n",
    " Average Precision  (AP) @[ IoU=0.50:0.95 | type=  easy | maxDets= 20 ] = 1.000\n",
    " Average Precision  (AP) @[ IoU=0.50:0.95 | type=medium | maxDets= 20 ] = 0.980\n",
    " Average Precision  (AP) @[ IoU=0.50:0.95 | type=  hard | maxDets= 20 ] = 0.412\n",
    "```"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}