diff --git a/tomato_model/working_tomato.ipynb b/tomato_model/working_tomato.ipynb
index 0ff48eb..5d213ab 100644
--- a/tomato_model/working_tomato.ipynb
+++ b/tomato_model/working_tomato.ipynb
@@ -1614,7 +1614,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -1624,13 +1624,135 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 3,
    "metadata": {},
    "outputs": [],
    "source": [
     "model_onnx = onnx.load(r\"runs\\detect\\train\\weights\\best.onnx\")"
    ]
   },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "onnx_model_path = \"runs/detect/train/weights/best.onnx\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import onnxruntime"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sess = onnxruntime.InferenceSession(onnx_model_path)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "img_path = \"val/images/IMG_20191215_112519.jpg\"\n",
+    "img = Image.open(img_path)\n",
+    "\n",
+    "# Resize the image to match the model input dimensions (640x640)\n",
+    "img = img.resize((640, 640))\n",
+    "\n",
+    "# Convert the image to a NumPy array and normalize if necessary\n",
+    "img_array = np.array(img).astype(np.float32) / 255.0\n",
+    "\n",
+    "# Add a batch dimension to match the model input shape\n",
+    "input_data = np.expand_dims(img_array, axis=0)\n",
+    "\n",
+    "# Transpose the array to match the expected shape [1, 3, 640, 640]\n",
+    "input_data = np.transpose(input_data, (0, 3, 1, 2))\n",
+    "\n",
+    "# Run inference\n",
+    "output = sess.run(None, {'images': input_data})\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib.patches as patches"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 640x480 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "output_array = output[0]\n",
+    "\n",
+    "# Extract the image from the array (assuming it represents an image)\n",
+    "image = np.squeeze(output_array)\n",
+    "\n",
+    "# Display the image using matplotlib\n",
+    "plt.imshow(image, cmap='viridis')  # You can choose a different colormap if needed\n",
+    "plt.colorbar()\n",
+    "plt.show()\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[array([[[     9.1795,      10.609,      15.407, ...,       571.7,      608.08,      637.42],\n",
+       "         [     14.536,      19.348,      13.814, ...,      640.77,      639.34,      633.99],\n",
+       "         [     18.398,      21.376,      31.429, ...,      387.02,      368.76,       333.3],\n",
+       "         ...,\n",
+       "         [ 1.3977e-05,  1.2845e-05,  8.1062e-06, ...,  1.7881e-07,  1.4901e-07,  1.7881e-07],\n",
+       "         [ 5.9545e-05,  4.3005e-05,  1.4186e-05, ...,  4.1723e-07,  3.2783e-07,  2.9802e-07],\n",
+       "         [ 0.00046247,  0.00027332,  6.2734e-05, ...,  5.3644e-07,  4.4703e-07,  4.7684e-07]]], dtype=float32)]"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "output"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 2,