docs: Add detailed comments to prediction notebook

Added comprehensive comments explaining the image preprocessing pipeline, model loading, and prediction logic to improve code documentation and understanding.

Author: arjun-ms

Model/prediction.ipynb

@@ -10,71 +10,37 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": null,
    "id": "c6bc823e",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "import numpy as np\n",
-    "import cv2\n",
-    "import keras\n",
-    "import os"
-   ]
+   "source": "# Import necessary libraries\nimport numpy as np  # For numerical operations and array handling\nimport cv2  # OpenCV for image processing\nimport keras  # For loading the trained model\nimport os  # For file path operations"
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "78b26d23",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "#preprocessing the user input\n",
-    "def image(path):\n",
-    "    img = cv2.imread(path, cv2.IMREAD_GRAYSCALE)\n",
-    "    new_arr = cv2.resize(img, (224, 224))\n",
-    "    new_arr = np.array(new_arr)/255\n",
-    "    new_arr = new_arr.reshape(-1, 224,224, 1)\n",
-    "    return new_arr"
-   ]
+   "source": "# Preprocessing function for user input images\ndef image(path):\n    # Read the image in grayscale mode\n    img = cv2.imread(path, cv2.IMREAD_GRAYSCALE)\n    \n    # Resize the image to 224x224 pixels to match model input requirements\n    new_arr = cv2.resize(img, (224, 224))\n    \n    # Normalize pixel values to range [0, 1] by dividing by 255\n    new_arr = np.array(new_arr)/255\n    \n    # Reshape to match model input shape: (batch_size, height, width, channels)\n    # -1 means infer batch size, 1 channel for grayscale\n    new_arr = new_arr.reshape(-1, 224, 224, 1)\n    \n    return new_arr"
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "id": "dc422667",
    "metadata": {},
    "outputs": [],
-   "source": [
-    "model = keras.models.load_model('../Server/3-class-improved.h5')"
-   ]
+   "source": "# Load the pre-trained model from the Server directory\nmodel = keras.models.load_model('../Server/3-class-improved.h5')"
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": null,
    "id": "f0421a50",
    "metadata": {
     "scrolled": true
    },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "mohanlal\n"
-     ]
-    }
-   ],
-   "source": [
-    "result = model.predict([image('test_images/mohanlal.png')])\n",
-    "\n",
-    "if result[0][0] > result[0][1] and result[0][0] > result[0][2]:\n",
-    "    prediction = 'Mammootty'\n",
-    "elif result[0][1] > result[0][0] and result[0][1] > result[0][2]:\n",
-    "    prediction = 'Mohanlal'\n",
-    "else:\n",
-    "    prediction = 'I dont know this guy!'\n",
-    "    \n",
-    "print(prediction)"
-   ]
+   "outputs": [],
+   "source": "# Get prediction probabilities for the test image\nresult = model.predict([image('test_images/mohanlal.png')])\n\n# Interpret the prediction results\n# Model outputs 3 probabilities: [Mammootty, Mohanlal, Unknown]\n# Compare probabilities to determine which class has the highest confidence\nif result[0][0] > result[0][1] and result[0][0] > result[0][2]:\n    prediction = 'Mammootty'\nelif result[0][1] > result[0][0] and result[0][1] > result[0][2]:\n    prediction = 'Mohanlal'\nelse:\n    # If the third class has highest probability, the person is unrecognized\n    prediction = 'I dont know this guy!'\n    \nprint(prediction)"
   },
   {
    "cell_type": "code",
@@ -111,4 +77,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}