handy/train/3_Test_Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Handy\n",
    "## Test model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "\n",
    "\n",
    "def calculate_angle_from_obj(a, b, c):\n",
    "    a = (a.x, a.y)  # First\n",
    "    b = (b.x, b.y)  # Mid - the angle corner\n",
    "    c = (c.x, c.y)  # End\n",
    "\n",
    "    return calculate_angle(a, b, c)\n",
    "\n",
    "\n",
    "def calculate_angle(a, b, c):\n",
    "    \"\"\"\n",
    "    Calculates the angle ABC given (x, y) of A, B and C \n",
    "\n",
    "    Args:\n",
    "        a (x, y)\n",
    "        b (x, y)\n",
    "        c (x, y)\n",
    "\n",
    "    Returns:\n",
    "        float: Angle (degrees)\n",
    "    \"\"\"\n",
    "    # Define points\n",
    "    A = np.array(a)\n",
    "    B = np.array(b)\n",
    "    C = np.array(c)\n",
    "\n",
    "    # Calculate vectors\n",
    "    BA = A - B\n",
    "    BC = C - B\n",
    "\n",
    "    # Calculate the cosine of the angle\n",
    "    cosine_angle = np.dot(BA, BC) / (np.linalg.norm(BA) * np.linalg.norm(BC))\n",
    "\n",
    "    # Calculate the angle and convert it to degrees\n",
    "    angle = np.arccos(cosine_angle)\n",
    "    angle_deg = np.degrees(angle)\n",
    "\n",
    "    return angle_deg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "File: 5_166.png\n",
      "Angles: 171.14, 1.11, 5.37, 147.35\n",
      "4\n",
      "[0.         0.         0.33333333 0.         0.63333333 0.03333333\n",
      " 0.         0.         0.         0.        ]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[7], line 85\u001b[0m\n\u001b[0;32m     83\u001b[0m         cv2\u001b[39m.\u001b[39mimshow(\u001b[39m\"\u001b[39m\u001b[39mHandy\u001b[39m\u001b[39m\"\u001b[39m, image)\n\u001b[0;32m     84\u001b[0m         \u001b[39mwhile\u001b[39;00m \u001b[39mTrue\u001b[39;00m:\n\u001b[1;32m---> 85\u001b[0m             \u001b[39mif\u001b[39;00m cv2\u001b[39m.\u001b[39;49mwaitKey(\u001b[39m1\u001b[39;49m) \u001b[39m&\u001b[39m \u001b[39m0xFF\u001b[39m \u001b[39m==\u001b[39m \u001b[39mord\u001b[39m(\u001b[39m\"\u001b[39m\u001b[39mq\u001b[39m\u001b[39m\"\u001b[39m):\n\u001b[0;32m     86\u001b[0m                 \u001b[39mbreak\u001b[39;00m\n\u001b[0;32m     87\u001b[0m cv2\u001b[39m.\u001b[39mdestroyAllWindows()\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "import os\n",
    "import random\n",
    "import cv2\n",
    "import numpy as np\n",
    "from os import path\n",
    "import mediapipe as mp\n",
    "import pickle\n",
    "\n",
    "import pandas as pd\n",
    "\n",
    "\n",
    "mp_holistic = mp.solutions.holistic\n",
    "mp_drawing = mp.solutions.drawing_utils\n",
    "\n",
    "model_path = \"handy_classifier.pkl\"\n",
    "\n",
    "\n",
    "if not path.exists(\"frames/\"):\n",
    "    print(\"Frames directory not found!\")\n",
    "    exit(-1)\n",
    "\n",
    "with mp_holistic.Holistic(\n",
    "        min_detection_confidence=0.5, min_tracking_confidence=0.5\n",
    "    ) as holistic, open(model_path, \"rb\") as f:\n",
    "        model = pickle.load(f)\n",
    "        items = os.listdir(\"frames/\")\n",
    "        items = [item for item in items if item.endswith(\".png\")]\n",
    "        item = random.choice(items)\n",
    "\n",
    "        print(f\"File: {item}\")\n",
    "\n",
    "        frame = cv2.imread(path.join(\"frames\", item))\n",
    "\n",
    "         # Recolor Feed\n",
    "        image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)\n",
    "        image.flags.writeable = False\n",
    "\n",
    "        # Make Detections\n",
    "        results = holistic.process(image)\n",
    "\n",
    "        if results.pose_landmarks is None:\n",
    "            print(\"Pose not detected\")\n",
    "            exit(-1)\n",
    "        \n",
    "        mp_drawing.draw_landmarks(\n",
    "            image,\n",
    "            results.pose_landmarks,\n",
    "            mp_holistic.POSE_CONNECTIONS,\n",
    "            mp_drawing.DrawingSpec(color=(245, 117, 66), thickness=1, circle_radius=2),\n",
    "            mp_drawing.DrawingSpec(color=(245, 66, 230), thickness=1, circle_radius=1),\n",
    "        )\n",
    "        # Calculate the angles\n",
    "        # See train/angles.png for more info\n",
    "\n",
    "        landmarks = results.pose_landmarks.landmark\n",
    "        angles = []\n",
    "        # Angle 0\n",
    "        angles.append(\n",
    "            calculate_angle_from_obj(landmarks[12], landmarks[14], landmarks[16])\n",
    "        )\n",
    "        # Angle 1\n",
    "        angles.append(\n",
    "            calculate_angle_from_obj(landmarks[11], landmarks[13], landmarks[15])\n",
    "        )\n",
    "        # Angle 2\n",
    "        angles.append(\n",
    "            calculate_angle_from_obj(landmarks[14], landmarks[12], landmarks[24])\n",
    "        )\n",
    "        # Angle 3\n",
    "        angles.append(\n",
    "            calculate_angle_from_obj(landmarks[13], landmarks[11], landmarks[23])\n",
    "        )\n",
    "\n",
    "        print(f\"Angles: {', '.join([f'{angle:.2f}' for angle in angles])}\")\n",
    "\n",
    "        # Try to make pose guess\n",
    "        X = pd.DataFrame([angles], columns=[f\"angle_{angle_index}\" for angle_index in range(len(angles))])\n",
    "        body_language_class = model.predict(X)[0]\n",
    "        body_language_prob = model.predict_proba(X)[0]\n",
    "        print(body_language_class)\n",
    "        print(body_language_prob)\n",
    "\n",
    "        cv2.imshow(\"Handy\", image)\n",
    "        while True:\n",
    "            if cv2.waitKey(1) & 0xFF == ord(\"q\"):\n",
    "                break\n",
    "cv2.destroyAllWindows()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "handy",
   "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": "3.11.4"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}