IQF_show.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Experiment Setup\n",
    "\n",
    "1. Define name of IQF experiment\n",
    "1. Define path of the original(reference) dataset\n",
    "1. Define path of the training script\n",
    "1. DS_wrapper is the class that encapsulate a dataset\n",
    "1. Define the list of modifications that will be applyed to the original dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "from iquaflow.datasets import DSModifier, DSWrapper,DSModifier_jpg\n",
    "from iquaflow.experiments import ExperimentInfo, ExperimentSetup\n",
    "from iquaflow.experiments.experiment_visual import ExperimentVisual\n",
    "from iquaflow.experiments.task_execution import PythonScriptTaskExecution\n",
    "from iquaflow.metrics import BBDetectionMetrics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment_name = \"experimentA\"\n",
    "data_path = \"/airport/dataset/SateAirports\"\n",
    "python_ml_script_path = \"./iqt_train_wrapper.py\"\n",
    "ds_wrapper = DSWrapper(data_path=data_path)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " * DSModifier refers to a modifier that simply does nothing - Reference training\n",
    " * DSModifier_jpg apply JPG compresion to the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds_modifiers_list = [ DSModifier_jpg(params={'quality': i}) for i in [10,30,50,70,90] ]\n",
    "task = PythonScriptTaskExecution( model_script_path=python_ml_script_path )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Experiment definition, pass as arguments all the components defined beforehand:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment = ExperimentSetup(\n",
    "    experiment_name=experiment_name,\n",
    "    task_instance=task,\n",
    "    ref_dsw_train=ds_wrapper,\n",
    "    ds_modifiers_list=ds_modifiers_list,\n",
    "    repetitions=5,\n",
    "    extra_train_params={'min_size':[11,16,20,30]}\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Execution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment.execute()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pick-up results / metrics and plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "\n",
    "from iquaflow.datasets import DSModifier, DSWrapper,DSModifier_jpg\n",
    "from iquaflow.experiments import ExperimentInfo, ExperimentSetup\n",
    "from iquaflow.experiments.experiment_visual import ExperimentVisual\n",
    "from iquaflow.experiments.task_execution import PythonScriptTaskExecution\n",
    "from iquaflow.metrics import BBDetectionMetrics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " * Define name of IQF experiment\n",
    " * Define path of the original(reference) dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "experiment_name = \"experimentA\"\n",
    "data_path = \"/airport/dataset/SateAirports\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load experiment info\n",
    " - ExperimentInfo is used to retrieve all the information of the whole experiment. \n",
    " - It contains built in operations but also it can be used to retrieve raw data for futher analysis\n",
    " - The get_df method will sort data in a dataframe suitable for the visualizations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "experiment_info = ExperimentInfo(experiment_name)\n",
    "\n",
    "df = experiment_info.get_df(\n",
    "    ds_params=[\"modifier\"],\n",
    "    metrics=[\"ap\",'ap_0.5_all','min_plane_size','quality'],\n",
    "    dropna = True,\n",
    "    fields_to_float_lst = ['ap','ap_0.5_all'],\n",
    "    fields_to_int_lst = ['quality','min_plane_size']\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>name</th>\n",
       "      <th>ds_modifier</th>\n",
       "      <th>ap</th>\n",
       "      <th>ap_0.5_all</th>\n",
       "      <th>min_plane_size</th>\n",
       "      <th>quality</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>SateAirports#jpg90_modifier</td>\n",
       "      <td>jpg90_modifier</td>\n",
       "      <td>0.936575</td>\n",
       "      <td>0.043528</td>\n",
       "      <td>30</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>SateAirports#jpg90_modifier_0</td>\n",
       "      <td>jpg90_modifier</td>\n",
       "      <td>0.887237</td>\n",
       "      <td>0.046832</td>\n",
       "      <td>30</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>SateAirports#jpg90_modifier_1</td>\n",
       "      <td>jpg90_modifier</td>\n",
       "      <td>0.874783</td>\n",
       "      <td>0.043926</td>\n",
       "      <td>30</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>SateAirports#jpg90_modifier_2</td>\n",
       "      <td>jpg90_modifier</td>\n",
       "      <td>0.879351</td>\n",
       "      <td>0.048211</td>\n",
       "      <td>30</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>SateAirports#jpg90_modifier_3</td>\n",
       "      <td>jpg90_modifier</td>\n",
       "      <td>0.888975</td>\n",
       "      <td>0.046086</td>\n",
       "      <td>30</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>SateAirports#jpg10_modifier_14</td>\n",
       "      <td>jpg10_modifier</td>\n",
       "      <td>0.520094</td>\n",
       "      <td>0.040593</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>SateAirports#jpg10_modifier_15</td>\n",
       "      <td>jpg10_modifier</td>\n",
       "      <td>0.507094</td>\n",
       "      <td>0.043096</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>SateAirports#jpg10_modifier_16</td>\n",
       "      <td>jpg10_modifier</td>\n",
       "      <td>0.542887</td>\n",
       "      <td>0.041962</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>SateAirports#jpg10_modifier_17</td>\n",
       "      <td>jpg10_modifier</td>\n",
       "      <td>0.489131</td>\n",
       "      <td>0.040315</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>SateAirports#jpg10_modifier_18</td>\n",
       "      <td>jpg10_modifier</td>\n",
       "      <td>0.504034</td>\n",
       "      <td>0.039628</td>\n",
       "      <td>11</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                              name     ds_modifier        ap  ap_0.5_all  \\\n",
       "0      SateAirports#jpg90_modifier  jpg90_modifier  0.936575    0.043528   \n",
       "1    SateAirports#jpg90_modifier_0  jpg90_modifier  0.887237    0.046832   \n",
       "2    SateAirports#jpg90_modifier_1  jpg90_modifier  0.874783    0.043926   \n",
       "3    SateAirports#jpg90_modifier_2  jpg90_modifier  0.879351    0.048211   \n",
       "4    SateAirports#jpg90_modifier_3  jpg90_modifier  0.888975    0.046086   \n",
       "..                             ...             ...       ...         ...   \n",
       "95  SateAirports#jpg10_modifier_14  jpg10_modifier  0.520094    0.040593   \n",
       "96  SateAirports#jpg10_modifier_15  jpg10_modifier  0.507094    0.043096   \n",
       "97  SateAirports#jpg10_modifier_16  jpg10_modifier  0.542887    0.041962   \n",
       "98  SateAirports#jpg10_modifier_17  jpg10_modifier  0.489131    0.040315   \n",
       "99  SateAirports#jpg10_modifier_18  jpg10_modifier  0.504034    0.039628   \n",
       "\n",
       "    min_plane_size  quality  \n",
       "0               30       90  \n",
       "1               30       90  \n",
       "2               30       90  \n",
       "3               30       90  \n",
       "4               30       90  \n",
       "..             ...      ...  \n",
       "95              11       10  \n",
       "96              11       10  \n",
       "97              11       10  \n",
       "98              11       10  \n",
       "99              11       10  \n",
       "\n",
       "[100 rows x 6 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualize\n",
    " - Plotting utilities are available in it_tool_box"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ev = ExperimentVisual(df, os.path.join(data_path, \"mod-ap.png\"))\n",
    "\n",
    "ev.visualize(\n",
    "    xvar=\"quality\",\n",
    "    yvar=\"ap\",\n",
    "    legend_var=\"min_plane_size\",\n",
    "    title=\"AP - quality\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ev = ExperimentVisual(df, os.path.join(data_path, \"mod-bars.png\"))\n",
    "\n",
    "ev.visualize(\n",
    "    plot_kind=\"bars\",\n",
    "    xvar=\"ds_modifier\",\n",
    "    yvar=\"ap\",\n",
    "    legend_var=\"min_plane_size\",\n",
    "    title=\"AP - quality\"\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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}