{
 "cells": [
  {
   "cell_type": "code",
   "id": "755f1267a6176c24",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-04-15T08:56:52.063622Z",
     "start_time": "2024-04-15T08:56:51.856029Z"
    }
   },
   "source": [
    "import json\n",
    "import pandas as pd"
   ],
   "execution_count": 1,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T08:57:05.764152Z",
     "start_time": "2024-04-15T08:57:05.390988Z"
    }
   },
   "cell_type": "code",
   "source": [
    "with open(\"./dataset/instances_attributes_val2020.json\", \"r\") as f:\n",
    "    dataset = json.load(f)\n",
    "\n",
    "print(dataset.keys())"
   ],
   "id": "initial_id",
   "execution_count": 3,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T09:41:18.117520Z",
     "start_time": "2024-04-15T09:41:18.110946Z"
    }
   },
   "cell_type": "code",
   "source": [
    "a = dataset['images']\n",
    "print(len(a))\n",
    "print(a)"
   ],
   "id": "24c0ff8dda428023",
   "execution_count": 30,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T12:48:49.207137Z",
     "start_time": "2024-04-15T12:48:49.202422Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df_categories = pd.DataFrame(dataset['categories'])\n",
    "df_categories"
   ],
   "id": "4e45dd93ed3d6d11",
   "execution_count": 107,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T09:02:58.490674Z",
     "start_time": "2024-04-15T09:02:58.485590Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df_attributes = pd.DataFrame(dataset['attributes'])\n",
    "df_attributes"
   ],
   "id": "f8c1f1649b4890b7",
   "execution_count": 18,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T09:02:21.405195Z",
     "start_time": "2024-04-15T09:02:21.390532Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df_annotation = pd.DataFrame(dataset['annotations'])\n",
    "df_annotation = df_annotation.drop(['segmentation'], axis='columns')\n",
    "df_annotation_exploded =df_annotation.explode('attribute_ids') "
   ],
   "id": "4c53c3115c3dbef8",
   "execution_count": 16,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T09:02:23.057097Z",
     "start_time": "2024-04-15T09:02:23.051760Z"
    }
   },
   "cell_type": "code",
   "source": "df_annotation_exploded",
   "id": "618f069b857225fc",
   "execution_count": 17,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T09:03:53.975762Z",
     "start_time": "2024-04-15T09:03:53.959100Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df_joined = df_attributes.merge(df_annotation_exploded, how='left', left_on='id', right_on='attribute_ids')\n",
    "df_joined"
   ],
   "id": "8a85bbf68e23920b",
   "execution_count": 20,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T09:08:05.647378Z",
     "start_time": "2024-04-15T09:08:05.637874Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df_neck_collar = df_joined[df_joined.name.str.contains('neck') | df_joined.name.str.contains('collar') ]\n",
    "df_neck_collar['name'].value_counts()"
   ],
   "id": "2888345540b560d5",
   "execution_count": 28,
   "outputs": []
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-04-15T09:49:34.287590Z",
     "start_time": "2024-04-15T09:49:34.263905Z"
    }
   },
   "cell_type": "code",
   "source": [
    "df_annotation = pd.DataFrame(dataset['annotations']).drop(['segmentation', 'attribute_ids'], axis='columns')\n",
    "df_images = pd.DataFrame(dataset['images'])\n",
    "df_categories = pd.DataFrame(dataset['categories'])\n",
    "\n",
    "df_dataset = pd.merge(df_images, df_annotation, how='inner', left_on='id', right_on='image_id')\n",
    "df_dataset = pd.merge(df_dataset, df_categories, how='inner', left_on='category_id', right_on='id')\n",
    "df_dataset['path_to_image'] = 'coucou/' + df_dataset['file_name']\n",
    "df_dataset"
   ],
   "id": "ebb14b7642a1459d",
   "execution_count": 37,
   "outputs": []
  },
  {
   "metadata": {},
   "cell_type": "code",
   "execution_count": null,
   "source": "",
   "id": "20ba29a56c257d56",
   "outputs": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}