plots_TS.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_n = pd.read_csv(\"our_predictions/prediction_006950_n.txt\")\n",
    "df_m = pd.read_csv(\"our_predictions/prediction_006950_m.txt\")\n",
    "df_x = pd.read_csv(\"our_predictions/prediction_006950_x.txt\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>img_path</th>\n",
       "      <th>inference_time</th>\n",
       "      <th>class_id</th>\n",
       "      <th>class_name</th>\n",
       "      <th>x_box</th>\n",
       "      <th>y_box</th>\n",
       "      <th>w_box</th>\n",
       "      <th>h_box</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>/work/vita/nmuenger_trinca/annotations/images/...</td>\n",
       "      <td>18.365383</td>\n",
       "      <td>154.0</td>\n",
       "      <td>F50.png</td>\n",
       "      <td>1122.455566</td>\n",
       "      <td>241.334671</td>\n",
       "      <td>56.266235</td>\n",
       "      <td>52.424286</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            img_path  inference_time  \\\n",
       "0  /work/vita/nmuenger_trinca/annotations/images/...       18.365383   \n",
       "\n",
       "   class_id class_name        x_box       y_box      w_box      h_box  \n",
       "0     154.0    F50.png  1122.455566  241.334671  56.266235  52.424286  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar([\"nano\", \"medium\", \"x-tra large\"],[df_n.iloc[0][\"inference_time\"],df_m.iloc[0][\"inference_time\"],df_x.iloc[0][\"inference_time\"]])\n",
    "plt.xlabel(\"Model name\")\n",
    "plt.ylabel(\"Inference time [ms]\")\n",
    "plt.title(\"Comparison of the models' inference time on one image\");\n",
    "plt.savefig(\"our_predictions/time_vs_model.png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.bar([\"nano\", \"medium\", \"x-tra large\"],[9602, 51786,133873])\n",
    "plt.xlabel(\"Model name\")\n",
    "plt.ylabel(\"Model size [Ko]\")\n",
    "plt.title(\"Comparison of the models' size\");\n",
    "plt.savefig(\"our_predictions/size_vs_model.png\")"
   ]
  }
 ],
 "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": "3.10.7"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}