1 line
No EOL
1.1 MiB
1 line
No EOL
1.1 MiB
{"cells":[{"cell_type":"markdown","id":"307bea88-e08c-4787-9d3a-047272601ef1","metadata":{"id":"307bea88-e08c-4787-9d3a-047272601ef1"},"source":["# Exploring and Analyzing Paris trees data"]},{"cell_type":"markdown","id":"39449094-4efa-4627-b19c-6aa521c7a7e8","metadata":{"id":"39449094-4efa-4627-b19c-6aa521c7a7e8"},"source":["Taken from https://s3-eu-west-1.amazonaws.com/static.oc-static.com/prod/courses/files/AI+Engineer/Project+2+Participez+%C3%A0+un+concours+sur+la+Smart+City/p2-arbres-fr.csv, coming from https://opendata.paris.fr/explore/dataset/les-arbres/map/?dataChart=eyJxdWVyaWVzIjpbeyJjb25maWciOnsiZGF0YXNldCI6Imxlcy1hcmJyZXMiLCJvcHRpb25zIjp7fX0sImNoYXJ0cyI6W3siYWxpZ25Nb250aCI6dHJ1ZSwidHlwZSI6ImNvbHVtbiIsImZ1bmMiOiJBVkciLCJ5QXhpcyI6ImlkYmFzZSIsInNjaWVudGlmaWNEaXNwbGF5Ijp0cnVlLCJjb2xvciI6IiMwMDMzNjYifV0sInhBeGlzIjoidHlwZWVtcGxhY2VtZW50IiwibWF4cG9pbnRzIjo1MCwic29ydCI6IiJ9XSwidGltZXNjYWxlIjoiIiwiZGlzcGxheUxlZ2VuZCI6dHJ1ZSwiYWxpZ25Nb250aCI6dHJ1ZX0%3D&disjunctive.typeemplacement&disjunctive.arrondissement&disjunctive.libellefrancais&disjunctive.genre&disjunctive.espece&disjunctive.varieteoucultivar&disjunctive.stadedeveloppement&disjunctive.remarquable&location=12,48.85984,2.31341&basemap=jawg.streets"]},{"cell_type":"markdown","id":"ceeb296c-0ca5-43d6-b5fb-a4620de1fc75","metadata":{"id":"ceeb296c-0ca5-43d6-b5fb-a4620de1fc75"},"source":["## 1) Opening csv file"]},{"cell_type":"code","execution_count":1,"id":"rA8K5-0Cmf33","metadata":{"id":"rA8K5-0Cmf33","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1737062546472,"user_tz":-60,"elapsed":2641,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"b849ecc7-fe2f-4c96-cd73-fcd37b003220"},"outputs":[{"output_type":"stream","name":"stdout","text":["Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n","Working directory: /content/drive/MyDrive/Colab_Notebooks/Project_2\n"]}],"source":["from google.colab import drive\n","drive.mount('/content/drive')\n","import os\n","os.listdir()\n","os.chdir(\"/content/drive/MyDrive/Colab_Notebooks/Project_2\")\n","print(f'Working directory: {os.getcwd()}')"]},{"cell_type":"code","execution_count":2,"id":"b4458a0a-853f-4a87-bb46-93ac6a9ba08b","metadata":{"id":"b4458a0a-853f-4a87-bb46-93ac6a9ba08b","executionInfo":{"status":"ok","timestamp":1737062547662,"user_tz":-60,"elapsed":1192,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["import pandas as pd\n","pd.set_option('display.float_format', '{:.2f}'.format) # don't want scientific notation. Floats are enough\n","\n","csv_path = \"./p2-trees-fr.csv\"\n","df = pd.read_csv(csv_path, sep=';')"]},{"cell_type":"markdown","id":"9de7fc9e-a309-46ac-a4f8-74effb1ca249","metadata":{"id":"9de7fc9e-a309-46ac-a4f8-74effb1ca249"},"source":["## 2) Naive Analysis"]},{"cell_type":"markdown","id":"d555a028-62b6-47ae-841b-13f45533ad3c","metadata":{"id":"d555a028-62b6-47ae-841b-13f45533ad3c"},"source":["### Use pandas basic methods"]},{"cell_type":"code","execution_count":3,"id":"5f0dc948-1c0f-4273-b8d5-7b9d0fc80f23","metadata":{"id":"5f0dc948-1c0f-4273-b8d5-7b9d0fc80f23","colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"status":"ok","timestamp":1737062548670,"user_tz":-60,"elapsed":1013,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"49cc84d0-f342-4904-b7fd-6fd0ed44f038"},"outputs":[{"output_type":"stream","name":"stdout","text":["Size in MB:\n"]},{"output_type":"execute_result","data":{"text/plain":["145.955268"]},"metadata":{},"execution_count":3}],"source":["import sys\n","print('Size in MB:')\n","sys.getsizeof(df)/(10**6)"]},{"cell_type":"code","execution_count":4,"id":"a302ef86-7553-4b38-8734-b0ae5896c584","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"a302ef86-7553-4b38-8734-b0ae5896c584","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062548671,"user_tz":-60,"elapsed":8,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"c072acc6-50e3-42ac-d905-a1369a0a6be6"},"outputs":[{"output_type":"stream","name":"stdout","text":["<class 'pandas.core.frame.DataFrame'>\n","RangeIndex: 200137 entries, 0 to 200136\n","Data columns (total 18 columns):\n"," # Column Non-Null Count Dtype \n","--- ------ -------------- ----- \n"," 0 id 200137 non-null int64 \n"," 1 type_emplacement 200137 non-null object \n"," 2 domanialite 200136 non-null object \n"," 3 arrondissement 200137 non-null object \n"," 4 complement_addresse 30902 non-null object \n"," 5 numero 0 non-null float64\n"," 6 lieu 200137 non-null object \n"," 7 id_emplacement 200137 non-null object \n"," 8 libelle_francais 198640 non-null object \n"," 9 genre 200121 non-null object \n"," 10 espece 198385 non-null object \n"," 11 variete 36777 non-null object \n"," 12 circonference_cm 200137 non-null int64 \n"," 13 hauteur_m 200137 non-null int64 \n"," 14 stade_developpement 132932 non-null object \n"," 15 remarquable 137039 non-null float64\n"," 16 geo_point_2d_a 200137 non-null float64\n"," 17 geo_point_2d_b 200137 non-null float64\n","dtypes: float64(4), int64(3), object(11)\n","memory usage: 27.5+ MB\n"]}],"source":["df.info() # display count of columns, lines and columns names"]},{"cell_type":"code","execution_count":5,"id":"f24becfa-7e0f-45a0-82b9-40e211922f5c","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":270},"id":"f24becfa-7e0f-45a0-82b9-40e211922f5c","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062548671,"user_tz":-60,"elapsed":6,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"ab0be1d5-2314-4f87-d4bf-a44da82aab5d"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" id type_emplacement domanialite arrondissement complement_addresse \\\n","0 99874 Arbre Jardin PARIS 7E ARRDT NaN \n","1 99875 Arbre Jardin PARIS 7E ARRDT NaN \n","\n"," numero lieu id_emplacement \\\n","0 NaN MAIRIE DU 7E 116 RUE DE GRENELLE PARIS 7E 19 \n","1 NaN MAIRIE DU 7E 116 RUE DE GRENELLE PARIS 7E 20 \n","\n"," libelle_francais genre espece variete circonference_cm \\\n","0 Marronnier Aesculus hippocastanum NaN 20 \n","1 If Taxus baccata NaN 65 \n","\n"," hauteur_m stade_developpement remarquable geo_point_2d_a geo_point_2d_b \n","0 5 NaN 0.00 48.86 2.32 \n","1 8 A NaN 48.86 2.32 "],"text/html":["\n"," <div id=\"df-f1f462f6-b0ad-4729-804d-04682c5cdcac\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>type_emplacement</th>\n"," <th>domanialite</th>\n"," <th>arrondissement</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>lieu</th>\n"," <th>id_emplacement</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>99874</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>MAIRIE DU 7E 116 RUE DE GRENELLE PARIS 7E</td>\n"," <td>19</td>\n"," <td>Marronnier</td>\n"," <td>Aesculus</td>\n"," <td>hippocastanum</td>\n"," <td>NaN</td>\n"," <td>20</td>\n"," <td>5</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>99875</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>MAIRIE DU 7E 116 RUE DE GRENELLE PARIS 7E</td>\n"," <td>20</td>\n"," <td>If</td>\n"," <td>Taxus</td>\n"," <td>baccata</td>\n"," <td>NaN</td>\n"," <td>65</td>\n"," <td>8</td>\n"," <td>A</td>\n"," <td>NaN</td>\n"," <td>48.86</td>\n"," <td>2.32</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-f1f462f6-b0ad-4729-804d-04682c5cdcac')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-f1f462f6-b0ad-4729-804d-04682c5cdcac button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-f1f462f6-b0ad-4729-804d-04682c5cdcac');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-25e1ad3f-0eb7-4603-adcc-98a9575a2829\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-25e1ad3f-0eb7-4603-adcc-98a9575a2829')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-25e1ad3f-0eb7-4603-adcc-98a9575a2829 button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"df"}},"metadata":{},"execution_count":5}],"source":["df.head(2) # display 2 first lines"]},{"cell_type":"code","execution_count":6,"id":"5543e915-5701-43ed-81dc-c1dd54715e95","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":300},"id":"5543e915-5701-43ed-81dc-c1dd54715e95","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062549223,"user_tz":-60,"elapsed":556,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"d6ad12a6-0590-4b03-c28e-f355a87f86dd"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" id numero circonference_cm hauteur_m remarquable \\\n","count 200137.00 0.00 200137.00 200137.00 137039.00 \n","mean 387202.68 NaN 83.38 13.11 0.00 \n","std 545603.24 NaN 673.19 1971.22 0.04 \n","min 99874.00 NaN 0.00 0.00 0.00 \n","25% 155927.00 NaN 30.00 5.00 0.00 \n","50% 221078.00 NaN 70.00 8.00 0.00 \n","75% 274102.00 NaN 115.00 12.00 0.00 \n","max 2024745.00 NaN 250255.00 881818.00 1.00 \n","\n"," geo_point_2d_a geo_point_2d_b \n","count 200137.00 200137.00 \n","mean 48.85 2.35 \n","std 0.03 0.05 \n","min 48.74 2.21 \n","25% 48.84 2.31 \n","50% 48.85 2.35 \n","75% 48.88 2.39 \n","max 48.91 2.47 "],"text/html":["\n"," <div id=\"df-d28dd5e2-6342-466c-b24a-8050035f8f9d\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>numero</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>count</th>\n"," <td>200137.00</td>\n"," <td>0.00</td>\n"," <td>200137.00</td>\n"," <td>200137.00</td>\n"," <td>137039.00</td>\n"," <td>200137.00</td>\n"," <td>200137.00</td>\n"," </tr>\n"," <tr>\n"," <th>mean</th>\n"," <td>387202.68</td>\n"," <td>NaN</td>\n"," <td>83.38</td>\n"," <td>13.11</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.35</td>\n"," </tr>\n"," <tr>\n"," <th>std</th>\n"," <td>545603.24</td>\n"," <td>NaN</td>\n"," <td>673.19</td>\n"," <td>1971.22</td>\n"," <td>0.04</td>\n"," <td>0.03</td>\n"," <td>0.05</td>\n"," </tr>\n"," <tr>\n"," <th>min</th>\n"," <td>99874.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>48.74</td>\n"," <td>2.21</td>\n"," </tr>\n"," <tr>\n"," <th>25%</th>\n"," <td>155927.00</td>\n"," <td>NaN</td>\n"," <td>30.00</td>\n"," <td>5.00</td>\n"," <td>0.00</td>\n"," <td>48.84</td>\n"," <td>2.31</td>\n"," </tr>\n"," <tr>\n"," <th>50%</th>\n"," <td>221078.00</td>\n"," <td>NaN</td>\n"," <td>70.00</td>\n"," <td>8.00</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.35</td>\n"," </tr>\n"," <tr>\n"," <th>75%</th>\n"," <td>274102.00</td>\n"," <td>NaN</td>\n"," <td>115.00</td>\n"," <td>12.00</td>\n"," <td>0.00</td>\n"," <td>48.88</td>\n"," <td>2.39</td>\n"," </tr>\n"," <tr>\n"," <th>max</th>\n"," <td>2024745.00</td>\n"," <td>NaN</td>\n"," <td>250255.00</td>\n"," <td>881818.00</td>\n"," <td>1.00</td>\n"," <td>48.91</td>\n"," <td>2.47</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-d28dd5e2-6342-466c-b24a-8050035f8f9d')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-d28dd5e2-6342-466c-b24a-8050035f8f9d button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-d28dd5e2-6342-466c-b24a-8050035f8f9d');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-df5a3df2-f99d-4f6b-b116-e2926db7147c\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-df5a3df2-f99d-4f6b-b116-e2926db7147c')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-df5a3df2-f99d-4f6b-b116-e2926db7147c button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n \"name\": \"## > 250000 cm as circumference, > 800000 m as height\",\n \"rows\": 8,\n \"fields\": [\n {\n \"column\": \"id\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 636490.3531323484,\n \"min\": 99874.0,\n \"max\": 2024745.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 387202.68230761925,\n 221078.0,\n 200137.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"circonference_cm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 105028.38312832116,\n \"min\": 0.0,\n \"max\": 250255.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 83.38047937163043\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"hauteur_m\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 309559.4466281352,\n \"min\": 0.0,\n \"max\": 881818.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 13.110509301128728\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 48450.55066982106,\n \"min\": 0.0,\n \"max\": 137039.0,\n \"num_unique_values\": 5,\n \"samples\": [\n 0.0013426834696692183\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_a\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 70744.31300132743,\n \"min\": 0.0302341454886933,\n \"max\": 200137.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 48.85449078901268\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_b\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 70758.40152298885,\n \"min\": 0.05121963963079915,\n \"max\": 200137.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 2.3482078955462966\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":6}],"source":["df.describe()# general statistics only on number types variables (int and floats)\n","\n","# problems:\n","## 0 as min value for circumference, height\n","## > 250000 cm as circumference, > 800000 m as height"]},{"cell_type":"code","execution_count":86,"id":"5c2225ff-7a7c-421c-9bde-1dc02e2e3742","metadata":{"id":"5c2225ff-7a7c-421c-9bde-1dc02e2e3742","executionInfo":{"status":"ok","timestamp":1737062771291,"user_tz":-60,"elapsed":423,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["from data_analysis_functions import plot_univariate_variable_analysis, plot_bivariate_variable_analysis, compute_bivariate_stats_indicators"]},{"cell_type":"code","source":["type(df[df['espece'].notna()]['espece'].values[0]) == str"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"la1Q1bZ0-B-L","executionInfo":{"status":"ok","timestamp":1737062552620,"user_tz":-60,"elapsed":5,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"c19a56cf-9388-4090-be52-b8e06ead8f84"},"id":"la1Q1bZ0-B-L","execution_count":9,"outputs":[{"output_type":"execute_result","data":{"text/plain":["True"]},"metadata":{},"execution_count":9}]},{"cell_type":"code","execution_count":10,"id":"f5aade17-79f8-4dfa-bd8f-547727c207cb","metadata":{"id":"f5aade17-79f8-4dfa-bd8f-547727c207cb","executionInfo":{"status":"ok","timestamp":1737062552620,"user_tz":-60,"elapsed":3,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["quantitative_variables = ['circonference_cm','hauteur_m']"]},{"cell_type":"markdown","id":"aec48b3b-3131-48b0-96a6-9d3a962162f1","metadata":{"id":"aec48b3b-3131-48b0-96a6-9d3a962162f1"},"source":["### Looking for data anomalies"]},{"cell_type":"markdown","id":"bd284b56-394b-4677-b271-91f374e04b01","metadata":{"id":"bd284b56-394b-4677-b271-91f374e04b01"},"source":["#### Outliers"]},{"cell_type":"code","execution_count":11,"id":"c0b79f60-4033-43ad-a8e2-7ba6022ea6f4","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":849},"id":"c0b79f60-4033-43ad-a8e2-7ba6022ea6f4","executionInfo":{"status":"ok","timestamp":1737062555058,"user_tz":-60,"elapsed":2441,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"be8a95c4-c24c-4ae5-f2c2-50b608f8eab7"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["# Boxplots in order to briefly find outliers\n","for variable in quantitative_variables:\n"," plot_univariate_variable_analysis(df, variable, 'boxplot')"]},{"cell_type":"markdown","id":"96db7b30-d700-47f0-950c-012817006a2b","metadata":{"id":"96db7b30-d700-47f0-950c-012817006a2b"},"source":["It seems that there are many outliers on data"]},{"cell_type":"markdown","id":"580de3b3-db0e-44cd-b271-2c619dc8dee4","metadata":{"id":"580de3b3-db0e-44cd-b271-2c619dc8dee4"},"source":["#### NaN"]},{"cell_type":"code","execution_count":12,"id":"cf0c8910-8bf2-4e28-a919-efacd17a0a2a","metadata":{"id":"cf0c8910-8bf2-4e28-a919-efacd17a0a2a","executionInfo":{"status":"ok","timestamp":1737062555059,"user_tz":-60,"elapsed":3,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["import missingno as msno\n","import numpy as np"]},{"cell_type":"code","execution_count":13,"id":"f9ba7bc9-8b7b-4203-8e59-6fbd7d0bc5cd","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"f9ba7bc9-8b7b-4203-8e59-6fbd7d0bc5cd","executionInfo":{"status":"ok","timestamp":1737062556368,"user_tz":-60,"elapsed":1313,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"9a36f968-1cae-48fa-e6e4-288cf963cecf"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([ 0, 0, 1, 0, 169235, 200137, 0, 0,\n"," 1497, 16, 1752, 163360, 0, 0, 67205, 63098,\n"," 0, 0])"]},"metadata":{},"execution_count":13}],"source":["# number of 'nan' for each column\n","column_nb_of_nans = np.array([], dtype='int')\n","for column in df.columns:\n"," where_nan = df[df[column].isna()]\n"," column_nb_of_nans = np.concatenate((column_nb_of_nans,[len(where_nan)]))\n","column_nb_of_nans"]},{"cell_type":"code","execution_count":14,"id":"4c0424ef-ae93-4b94-847c-4c8341051aa6","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":107},"id":"4c0424ef-ae93-4b94-847c-4c8341051aa6","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062556798,"user_tz":-60,"elapsed":434,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"36154d15-5ffd-48e2-f046-632428de3cef"},"outputs":[{"output_type":"stream","name":"stdout","text":["Number of nan for each variable that contains minimum 1 occurrence:\n"]},{"output_type":"execute_result","data":{"text/plain":[" domanialite complement_addresse numero libelle_francais genre espece \\\n","0 1 169235 200137 1497 16 1752 \n","\n"," variete stade_developpement remarquable \n","0 163360 67205 63098 "],"text/html":["\n"," <div id=\"df-407b1e60-41d7-4d7f-a104-a11eb4268da6\" class=\"colab-df-container\">\n"," <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>domanialite</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1</td>\n"," <td>169235</td>\n"," <td>200137</td>\n"," <td>1497</td>\n"," <td>16</td>\n"," <td>1752</td>\n"," <td>163360</td>\n"," <td>67205</td>\n"," <td>63098</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-407b1e60-41d7-4d7f-a104-a11eb4268da6')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-407b1e60-41d7-4d7f-a104-a11eb4268da6 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-407b1e60-41d7-4d7f-a104-a11eb4268da6');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n"," <div id=\"id_90cd5411-0f7f-4fac-9933-565e39f6ab00\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('columns_with_nans_df')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_90cd5411-0f7f-4fac-9933-565e39f6ab00 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('columns_with_nans_df');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"columns_with_nans_df","summary":"{\n \"name\": \"columns_with_nans_df\",\n \"rows\": 1,\n \"fields\": [\n {\n \"column\": \"domanialite\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 1,\n \"max\": 1,\n \"num_unique_values\": 1,\n \"samples\": [\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"complement_addresse\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 169235,\n \"max\": 169235,\n \"num_unique_values\": 1,\n \"samples\": [\n 169235\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 200137,\n \"max\": 200137,\n \"num_unique_values\": 1,\n \"samples\": [\n 200137\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 1497,\n \"max\": 1497,\n \"num_unique_values\": 1,\n \"samples\": [\n 1497\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 16,\n \"max\": 16,\n \"num_unique_values\": 1,\n \"samples\": [\n 16\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 1752,\n \"max\": 1752,\n \"num_unique_values\": 1,\n \"samples\": [\n 1752\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 163360,\n \"max\": 163360,\n \"num_unique_values\": 1,\n \"samples\": [\n 163360\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 67205,\n \"max\": 67205,\n \"num_unique_values\": 1,\n \"samples\": [\n 67205\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 63098,\n \"max\": 63098,\n \"num_unique_values\": 1,\n \"samples\": [\n 63098\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":14}],"source":["# dataframe with number of nans per column only when one nan or more\n","where_nans = np.where(column_nb_of_nans>0)\n","columns_with_nans_df = pd.DataFrame([column_nb_of_nans[where_nans]], columns = df.columns[where_nans])\n","print('Number of nan for each variable that contains minimum 1 occurrence:')\n","columns_with_nans_df"]},{"cell_type":"code","execution_count":15,"id":"6d1dbd66-cc76-4724-9a23-eeefb13bf725","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":610},"id":"6d1dbd66-cc76-4724-9a23-eeefb13bf725","executionInfo":{"status":"ok","timestamp":1737062561949,"user_tz":-60,"elapsed":5152,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"e3b9d24a-db1c-48b4-a9a3-ae1b62fbba54"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["<Axes: >"]},"metadata":{},"execution_count":15},{"output_type":"display_data","data":{"text/plain":["<Figure size 2500x1000 with 3 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["# stade_developpement and remarquable are missing around 30% values. Variete is missing 80 percent values\n","msno.bar(df)"]},{"cell_type":"code","execution_count":16,"id":"ef1eb4db-5bfb-49a2-a4f9-cdc5249ad271","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":796},"id":"ef1eb4db-5bfb-49a2-a4f9-cdc5249ad271","executionInfo":{"status":"ok","timestamp":1737062564994,"user_tz":-60,"elapsed":3047,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"3ac44711-3e87-4027-e8c1-f08dc7bfbaa2"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["<Axes: >"]},"metadata":{},"execution_count":16},{"output_type":"display_data","data":{"text/plain":["<Figure size 2000x1200 with 2 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["# missing values correlation between stade_developpement and remarquable. Look at which trees they are\n","msno.heatmap(df)"]},{"cell_type":"markdown","id":"e173bf2d-060c-4610-bbf1-ef35f4a20552","metadata":{"id":"e173bf2d-060c-4610-bbf1-ef35f4a20552"},"source":["## 3) Cleaning anomalies"]},{"cell_type":"code","execution_count":17,"id":"57fe3e6a-3828-4f45-bede-a3422cff30c3","metadata":{"id":"57fe3e6a-3828-4f45-bede-a3422cff30c3","executionInfo":{"status":"ok","timestamp":1737062564994,"user_tz":-60,"elapsed":15,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["df_refined = df.copy()"]},{"cell_type":"markdown","id":"c5535b2a-fd5e-4b16-a1d3-bdc974e55dd0","metadata":{"id":"c5535b2a-fd5e-4b16-a1d3-bdc974e55dd0"},"source":["### Outliers (min and max)"]},{"cell_type":"markdown","id":"76dfdc46-4b19-4f63-a854-82f07a168921","metadata":{"id":"76dfdc46-4b19-4f63-a854-82f07a168921"},"source":["##### Max"]},{"cell_type":"markdown","id":"eeee8c4a-af66-4be0-9637-2054e67e7af5","metadata":{"id":"eeee8c4a-af66-4be0-9637-2054e67e7af5"},"source":["Height :"]},{"cell_type":"code","execution_count":18,"id":"ac8fcbb6-dafa-4f63-9caa-36f99a3a9297","metadata":{"id":"ac8fcbb6-dafa-4f63-9caa-36f99a3a9297","executionInfo":{"status":"ok","timestamp":1737062564994,"user_tz":-60,"elapsed":15,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# based on the tallest tree height : https://en.wikipedia.org/wiki/Hyperion_(tree)\n","# max_height = 116 # in meters\n","# tallest tree in Paris is supposed to be this one (40 m), so the others heights are wrong : https://france3-regions.francetvinfo.fr/paris-ile-de-france/paris/mais-quel-est-le-plus-grand-arbre-de-paris-2982575.html\n","max_height = 45 # in meters"]},{"cell_type":"code","execution_count":19,"id":"c3115746-c299-495f-8a13-b1e3e5ebc6cd","metadata":{"id":"c3115746-c299-495f-8a13-b1e3e5ebc6cd","executionInfo":{"status":"ok","timestamp":1737062564994,"user_tz":-60,"elapsed":14,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["max_height_outlier_df = df_refined[(df_refined['hauteur_m'] >= max_height)]"]},{"cell_type":"code","execution_count":20,"id":"ba6ca618-9335-4f54-8215-1545e821e3da","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":895},"id":"ba6ca618-9335-4f54-8215-1545e821e3da","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062564994,"user_tz":-60,"elapsed":14,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"c3b39980-eb60-4b6a-c95d-2856c063c926"},"outputs":[{"output_type":"stream","name":"stdout","text":["height outliers trees (479 occurrencies):\n"]},{"output_type":"execute_result","data":{"text/plain":[" id type_emplacement domanialite arrondissement \\\n","4534 105885 Arbre Jardin PARIS 13E ARRDT \n","4537 105888 Arbre Jardin PARIS 13E ARRDT \n","5392 106800 Arbre Jardin PARIS 7E ARRDT \n","5469 106880 Arbre Jardin PARIS 7E ARRDT \n","5599 107029 Arbre Jardin PARIS 7E ARRDT \n","... ... ... ... ... \n","194812 2018093 Arbre Alignement PARIS 18E ARRDT \n","197026 2020675 Arbre Jardin PARIS 18E ARRDT \n","197060 2020711 Arbre Alignement PARIS 19E ARRDT \n","197136 2020792 Arbre Jardin PARIS 14E ARRDT \n","197146 2020802 Arbre Jardin PARIS 14E ARRDT \n","\n"," complement_addresse numero \\\n","4534 NaN NaN \n","4537 NaN NaN \n","5392 NaN NaN \n","5469 NaN NaN \n","5599 NaN NaN \n","... ... ... \n","194812 95 NaN \n","197026 NaN NaN \n","197060 NaN NaN \n","197136 NaN NaN \n","197146 NaN NaN \n","\n"," lieu id_emplacement \\\n","4534 SQUARE DE CHOISY 277 \n","4537 SQUARE DE CHOISY 280 \n","5392 JARDIN DU CHAMP DE MARS A0120683 \n","5469 JARDIN DU CHAMP DE MARS P0070763 \n","5599 JARDIN DU CHAMP DE MARS A0041225 \n","... ... ... \n","194812 RUE DE LA CHAPELLE 1501013 \n","197026 JARDINIERES T3 BOULEVARD NEY CHAPELLE / 2 AVEN... 101016 \n","197060 QUAI DE LA CHARENTE 201004 \n","197136 SQUARE AUGUSTE RENOIR / 205 RUE RAYMOND LOSSERAND 103005 \n","197146 SQUARE AUGUSTE RENOIR / 205 RUE RAYMOND LOSSERAND 104004 \n","\n"," libelle_francais genre espece variete circonference_cm \\\n","4534 Fevier Gleditsia triacanthos NaN 45 \n","4537 Tulipier Liriodendron tulipifera NaN 45 \n","5392 Tilleul Tilia n. sp. NaN 100 \n","5469 If Taxus baccata NaN 61 \n","5599 Platane Platanus x hispanica NaN 41 \n","... ... ... ... ... ... \n","194812 Platane Platanus x hispanica NaN 35 \n","197026 Pin Pinus nigra NaN 50 \n","197060 Aulne Alnus cordata NaN 30 \n","197136 Erable Acer platanoides NaN 145 \n","197146 Erable Acer platanoides NaN 150 \n","\n"," hauteur_m stade_developpement remarquable geo_point_2d_a \\\n","4534 45 NaN NaN 48.83 \n","4537 45 NaN NaN 48.83 \n","5392 1510 A 0.00 48.86 \n","5469 65 A 0.00 48.85 \n","5599 68 JA 0.00 48.86 \n","... ... ... ... ... \n","194812 65 J 0.00 48.90 \n","197026 65 J 0.00 48.90 \n","197060 55 JA 0.00 48.90 \n","197136 125 A 0.00 48.83 \n","197146 105 A 0.00 48.83 \n","\n"," geo_point_2d_b \n","4534 2.36 \n","4537 2.36 \n","5392 2.30 \n","5469 2.30 \n","5599 2.30 \n","... ... \n","194812 2.36 \n","197026 2.36 \n","197060 2.38 \n","197136 2.31 \n","197146 2.31 \n","\n","[479 rows x 18 columns]"],"text/html":["\n"," <div id=\"df-a84a1757-f7bc-4f23-82ac-ad835b448e01\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>type_emplacement</th>\n"," <th>domanialite</th>\n"," <th>arrondissement</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>lieu</th>\n"," <th>id_emplacement</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>4534</th>\n"," <td>105885</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 13E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>SQUARE DE CHOISY</td>\n"," <td>277</td>\n"," <td>Fevier</td>\n"," <td>Gleditsia</td>\n"," <td>triacanthos</td>\n"," <td>NaN</td>\n"," <td>45</td>\n"," <td>45</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.83</td>\n"," <td>2.36</td>\n"," </tr>\n"," <tr>\n"," <th>4537</th>\n"," <td>105888</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 13E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>SQUARE DE CHOISY</td>\n"," <td>280</td>\n"," <td>Tulipier</td>\n"," <td>Liriodendron</td>\n"," <td>tulipifera</td>\n"," <td>NaN</td>\n"," <td>45</td>\n"," <td>45</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.83</td>\n"," <td>2.36</td>\n"," </tr>\n"," <tr>\n"," <th>5392</th>\n"," <td>106800</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>JARDIN DU CHAMP DE MARS</td>\n"," <td>A0120683</td>\n"," <td>Tilleul</td>\n"," <td>Tilia</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," <td>100</td>\n"," <td>1510</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.30</td>\n"," </tr>\n"," <tr>\n"," <th>5469</th>\n"," <td>106880</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>JARDIN DU CHAMP DE MARS</td>\n"," <td>P0070763</td>\n"," <td>If</td>\n"," <td>Taxus</td>\n"," <td>baccata</td>\n"," <td>NaN</td>\n"," <td>61</td>\n"," <td>65</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.30</td>\n"," </tr>\n"," <tr>\n"," <th>5599</th>\n"," <td>107029</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>JARDIN DU CHAMP DE MARS</td>\n"," <td>A0041225</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>41</td>\n"," <td>68</td>\n"," <td>JA</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.30</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"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>194812</th>\n"," <td>2018093</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 18E ARRDT</td>\n"," <td>95</td>\n"," <td>NaN</td>\n"," <td>RUE DE LA CHAPELLE</td>\n"," <td>1501013</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>35</td>\n"," <td>65</td>\n"," <td>J</td>\n"," <td>0.00</td>\n"," <td>48.90</td>\n"," <td>2.36</td>\n"," </tr>\n"," <tr>\n"," <th>197026</th>\n"," <td>2020675</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 18E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>JARDINIERES T3 BOULEVARD NEY CHAPELLE / 2 AVEN...</td>\n"," <td>101016</td>\n"," <td>Pin</td>\n"," <td>Pinus</td>\n"," <td>nigra</td>\n"," <td>NaN</td>\n"," <td>50</td>\n"," <td>65</td>\n"," <td>J</td>\n"," <td>0.00</td>\n"," <td>48.90</td>\n"," <td>2.36</td>\n"," </tr>\n"," <tr>\n"," <th>197060</th>\n"," <td>2020711</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 19E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>QUAI DE LA CHARENTE</td>\n"," <td>201004</td>\n"," <td>Aulne</td>\n"," <td>Alnus</td>\n"," <td>cordata</td>\n"," <td>NaN</td>\n"," <td>30</td>\n"," <td>55</td>\n"," <td>JA</td>\n"," <td>0.00</td>\n"," <td>48.90</td>\n"," <td>2.38</td>\n"," </tr>\n"," <tr>\n"," <th>197136</th>\n"," <td>2020792</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 14E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>SQUARE AUGUSTE RENOIR / 205 RUE RAYMOND LOSSERAND</td>\n"," <td>103005</td>\n"," <td>Erable</td>\n"," <td>Acer</td>\n"," <td>platanoides</td>\n"," <td>NaN</td>\n"," <td>145</td>\n"," <td>125</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.83</td>\n"," <td>2.31</td>\n"," </tr>\n"," <tr>\n"," <th>197146</th>\n"," <td>2020802</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 14E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>SQUARE AUGUSTE RENOIR / 205 RUE RAYMOND LOSSERAND</td>\n"," <td>104004</td>\n"," <td>Erable</td>\n"," <td>Acer</td>\n"," <td>platanoides</td>\n"," <td>NaN</td>\n"," <td>150</td>\n"," <td>105</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.83</td>\n"," <td>2.31</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>479 rows × 18 columns</p>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a84a1757-f7bc-4f23-82ac-ad835b448e01')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-a84a1757-f7bc-4f23-82ac-ad835b448e01 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-a84a1757-f7bc-4f23-82ac-ad835b448e01');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-0fe2d811-c07a-4325-a1f8-b36c7f650043\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-0fe2d811-c07a-4325-a1f8-b36c7f650043')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-0fe2d811-c07a-4325-a1f8-b36c7f650043 button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_7f99ae72-76b6-48d7-b5c9-43062bc04af0\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('max_height_outlier_df')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_7f99ae72-76b6-48d7-b5c9-43062bc04af0 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('max_height_outlier_df');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"max_height_outlier_df","summary":"{\n \"name\": \"max_height_outlier_df\",\n \"rows\": 479,\n \"fields\": [\n {\n \"column\": \"id\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 887737,\n \"min\": 105885,\n \"max\": 2020802,\n \"num_unique_values\": 479,\n \"samples\": [\n 226659,\n 2008767,\n 202895\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"type_emplacement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Arbre\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"domanialite\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 7,\n \"samples\": [\n \"Jardin\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"arrondissement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 22,\n \"samples\": [\n \"PARIS 13E ARRDT\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"complement_addresse\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 56,\n \"samples\": [\n \"11 \\u00e0 27\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": null,\n \"max\": null,\n \"num_unique_values\": 0,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lieu\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 177,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"id_emplacement\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 440,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 39,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 36,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 57,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 19,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"circonference_cm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 428,\n \"min\": 0,\n \"max\": 7540,\n \"num_unique_values\": 80,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"hauteur_m\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 40285,\n \"min\": 45,\n \"max\": 881818,\n \"num_unique_values\": 102,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 4,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0654416970922002,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 2,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_a\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.025974849186796975,\n \"min\": 48.810646078,\n \"max\": 48.9089446402,\n \"num_unique_values\": 479,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_b\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.035886893415735685,\n \"min\": 2.2528752018,\n \"max\": 2.4613206192,\n \"num_unique_values\": 479,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":20}],"source":["print('height outliers trees ('+ str(len(max_height_outlier_df)) + ' occurrencies):')\n","max_height_outlier_df\n"]},{"cell_type":"code","execution_count":21,"id":"bf489b0f-45f5-4879-8c56-b69f6b79f89e","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":489},"id":"bf489b0f-45f5-4879-8c56-b69f6b79f89e","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062564994,"user_tz":-60,"elapsed":11,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"88793feb-60c1-4363-f936-cf7a43dae03a"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["hauteur_m\n","45 19\n","46 1\n","47 1\n","50 13\n","52 1\n"," ..\n","1815 1\n","2015 1\n","2525 1\n","5155 1\n","881818 1\n","Name: count, Length: 102, dtype: int64"],"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>count</th>\n"," </tr>\n"," <tr>\n"," <th>hauteur_m</th>\n"," <th></th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>45</th>\n"," <td>19</td>\n"," </tr>\n"," <tr>\n"," <th>46</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>47</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>50</th>\n"," <td>13</td>\n"," </tr>\n"," <tr>\n"," <th>52</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>1815</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>2015</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>2525</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>5155</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>881818</th>\n"," <td>1</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>102 rows × 1 columns</p>\n","</div><br><label><b>dtype:</b> int64</label>"]},"metadata":{},"execution_count":21}],"source":["max_height_outlier_df.value_counts('hauteur_m').sort_index(ascending=True)"]},{"cell_type":"code","execution_count":22,"id":"c788b243-2521-4f58-8d1f-3463ec0bf45c","metadata":{"id":"c788b243-2521-4f58-8d1f-3463ec0bf45c","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062564994,"user_tz":-60,"elapsed":9,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# remove outlier value\n","df_refined.loc[df_refined['hauteur_m'] == 881818, 'hauteur_m'] = np.nan"]},{"cell_type":"code","source":["for variable in quantitative_variables:\n"," plot_univariate_variable_analysis(df_refined, variable, 'boxplot')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":849},"id":"K9ZrNsYvsDxk","executionInfo":{"status":"ok","timestamp":1737062566942,"user_tz":-60,"elapsed":1956,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"408da53e-c277-4dff-f93a-e068b303f6b3"},"id":"K9ZrNsYvsDxk","execution_count":23,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","execution_count":24,"id":"76b470a5-7542-44a0-b262-fd9e325b4a43","metadata":{"id":"76b470a5-7542-44a0-b262-fd9e325b4a43","executionInfo":{"status":"ok","timestamp":1737062566942,"user_tz":-60,"elapsed":67,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# need in order to divide by 100 the too high heights\n","df_refined['hauteur_m'] = df_refined['hauteur_m'].astype(float)"]},{"cell_type":"code","execution_count":25,"id":"e0cffc30-00b3-439c-9501-9e94b1f57e0c","metadata":{"id":"e0cffc30-00b3-439c-9501-9e94b1f57e0c","executionInfo":{"status":"ok","timestamp":1737062566943,"user_tz":-60,"elapsed":66,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# Too high remaining heights may have been measured in cms\n","df_refined.loc[df_refined['hauteur_m'] >= max_height, 'hauteur_m'] = df_refined.loc[df_refined['hauteur_m'] >= max_height, 'hauteur_m']/100"]},{"cell_type":"code","execution_count":26,"id":"TVbNwJ0DF9zQ","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"TVbNwJ0DF9zQ","executionInfo":{"status":"ok","timestamp":1737062566943,"user_tz":-60,"elapsed":65,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"af8b9f10-e138-4a17-9d4b-f16e17bcd042"},"outputs":[{"output_type":"stream","name":"stdout","text":["Amount of to high heights : 479.\n","1 value has been removed (set to 'NaN').\n","478 values may have been measured in cms --> converted in meters.\n"]}],"source":["print(f\"Amount of to high heights : {len(max_height_outlier_df)}.\\n1 value has been removed (set to 'NaN').\\n{len(max_height_outlier_df)-1} values may have been measured in cms --> converted in meters.\")"]},{"cell_type":"markdown","id":"1e751c26-fc8f-41a9-bd27-2d1f899f4866","metadata":{"id":"1e751c26-fc8f-41a9-bd27-2d1f899f4866"},"source":["Girth :"]},{"cell_type":"code","execution_count":27,"id":"0b2af1bd-9fc8-405f-b810-433044a0ea7a","metadata":{"id":"0b2af1bd-9fc8-405f-b810-433044a0ea7a","executionInfo":{"status":"ok","timestamp":1737062566943,"user_tz":-60,"elapsed":63,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# based on the biggest tree girth (circumference) : https://www.guinnessworldrecords.com/world-records/living-tree-with-greatest-girth\n","max_girth = 36 # in meters"]},{"cell_type":"code","execution_count":28,"id":"dceb4f45-c91d-4234-a5d1-c87d1fcc6403","metadata":{"id":"dceb4f45-c91d-4234-a5d1-c87d1fcc6403","executionInfo":{"status":"ok","timestamp":1737062566943,"user_tz":-60,"elapsed":63,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["max_girth_outlier_df = df_refined[(df_refined['circonference_cm']/100 > max_girth)]"]},{"cell_type":"code","execution_count":29,"id":"a97d6d26-f82a-463b-ad00-69d4676beb7e","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"collapsed":true,"id":"a97d6d26-f82a-463b-ad00-69d4676beb7e","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062566943,"user_tz":-60,"elapsed":62,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"8aedd841-8816-45e4-a550-e7130a0e7a67"},"outputs":[{"output_type":"stream","name":"stdout","text":["girth outliers trees (20 occurrencies):\n"]},{"output_type":"execute_result","data":{"text/plain":[" id type_emplacement domanialite arrondissement \\\n","71787 190080 Arbre DASCO PARIS 20E ARRDT \n","82951 203295 Arbre Alignement PARIS 18E ARRDT \n","90605 211362 Arbre Alignement PARIS 5E ARRDT \n","91256 212024 Arbre Alignement PARIS 12E ARRDT \n","97354 218274 Arbre Alignement PARIS 7E ARRDT \n","103437 224609 Arbre Alignement PARIS 18E ARRDT \n","133546 256868 Arbre Alignement PARIS 19E ARRDT \n","147235 270997 Arbre Alignement PARIS 7E ARRDT \n","147634 271402 Arbre Alignement PARIS 7E ARRDT \n","148106 271884 Arbre Alignement PARIS 7E ARRDT \n","148481 272263 Arbre Alignement PARIS 7E ARRDT \n","149056 272885 Arbre Alignement PARIS 18E ARRDT \n","151364 275451 Arbre Alignement PARIS 7E ARRDT \n","152303 276396 Arbre Alignement PARIS 7E ARRDT \n","160432 284823 Arbre Alignement PARIS 18E ARRDT \n","163918 288407 Arbre Alignement PARIS 12E ARRDT \n","164616 289121 Arbre Alignement PARIS 12E ARRDT \n","167606 292180 Arbre Alignement PARIS 18E ARRDT \n","170367 295047 Arbre Alignement PARIS 5E ARRDT \n","173343 298211 Arbre Alignement PARIS 5E ARRDT \n","\n"," complement_addresse numero lieu \\\n","71787 NaN NaN PLAINE (9) ELEM B \n","82951 NaN NaN RUE D AUBERVILLIERS \n","90605 34 NaN BOULEVARD SAINT MARCEL \n","91256 NaN NaN RUE DE CHALON \n","97354 27 NaN AVENUE RAPP \n","103437 NaN NaN BOULEVARD BARBES \n","133546 NaN NaN AVENUE JEAN JAURES \n","147235 NaN NaN QUAI BRANLY \n","147634 NaN NaN AVENUE BOSQUET \n","148106 NaN NaN AVENUE BOSQUET \n","148481 14 NaN BOULEVARD DES INVALIDES \n","149056 NaN NaN PLACE DE LA CHAPELLE \n","151364 NaN NaN ESPLANADE DES INVALIDES \n","152303 NaN NaN ESPLANADE DES INVALIDES \n","160432 NaN NaN AVENUE DE LA PORTE DE CLIGNANCOURT \n","163918 NaN NaN BOULEVARD DE PICPUS \n","164616 NaN NaN RUE FRANCOIS TRUFFAUT \n","167606 16 NaN RUE JEAN COCTEAU \n","170367 16 NaN RUE POLIVEAU \n","173343 NaN NaN BOULEVARD DE PORT ROYAL \n","\n"," id_emplacement libelle_francais genre espece \\\n","71787 11110006 Tilleul Tilia tomentosa \n","82951 301004 Erable Acer monspessulanum \n","90605 204025 Cedrele Toona sinensis \n","91256 202005 Micocoulier Celtis australis \n","97354 501003 Platane Platanus occidentalis \n","103437 1303003 Platane Platanus x hispanica \n","133546 1501006 Platane Platanus x hispanica \n","147235 503005 Paulownia Paulownia tomentosa \n","147634 1202016 Platane Platanus occidentalis \n","148106 501022 Platane Platanus occidentalis \n","148481 402009 Platane Platanus occidentalis \n","149056 101026 Platane Platanus x hispanica \n","151364 410005 Tilleul Tilia tomentosa \n","152303 402011 Tilleul Tilia tomentosa \n","160432 202005 Platane Platanus orientalis \n","163918 1604017 Sophora Sophora japonica \n","164616 301005 Platane Platanus x hispanica \n","167606 202027 Noisetier de Byzance Corylus colurna \n","170367 202015 Platane Platanus x hispanica \n","173343 402001 Marronnier Aesculus x carnea \n","\n"," variete circonference_cm hauteur_m stade_developpement \\\n","71787 NaN 18589 15.00 A \n","82951 NaN 5545 0.56 J \n","90605 NaN 20130 16.00 A \n","91256 NaN 30140 10.00 A \n","97354 NaN 8110 18.00 A \n","103437 NaN 9035 6.00 J \n","133546 NaN 4090 10.00 A \n","147235 NaN 9155 15.00 A \n","147634 NaN 65110 16.00 A \n","148106 NaN 6595 15.00 A \n","148481 NaN 6120 18.00 A \n","149056 NaN 250255 22.00 A \n","151364 NaN 7540 8.15 JA \n","152303 NaN 9085 12.00 A \n","160432 NaN 120170 20.00 A \n","163918 NaN 6085 5.00 JA \n","164616 NaN 9095 10.00 A \n","167606 NaN 3630 5.00 J \n","170367 NaN 6125 20.00 A \n","173343 Briotii' 79120 10.00 A \n","\n"," remarquable geo_point_2d_a geo_point_2d_b \n","71787 0.00 48.85 2.40 \n","82951 0.00 48.89 2.37 \n","90605 0.00 48.84 2.36 \n","91256 0.00 48.85 2.37 \n","97354 0.00 48.86 2.30 \n","103437 0.00 48.89 2.35 \n","133546 0.00 48.89 2.39 \n","147235 0.00 48.86 2.29 \n","147634 0.00 48.86 2.31 \n","148106 0.00 48.86 2.30 \n","148481 0.00 48.85 2.31 \n","149056 0.00 48.88 2.36 \n","151364 0.00 48.86 2.31 \n","152303 0.00 48.86 2.31 \n","160432 0.00 48.90 2.34 \n","163918 0.00 48.85 2.40 \n","164616 0.00 48.83 2.39 \n","167606 0.00 48.90 2.35 \n","170367 0.00 48.84 2.36 \n","173343 0.00 48.84 2.35 "],"text/html":["\n"," <div id=\"df-5076a33a-6a75-4619-8a9b-6c434233f810\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>type_emplacement</th>\n"," <th>domanialite</th>\n"," <th>arrondissement</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>lieu</th>\n"," <th>id_emplacement</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>71787</th>\n"," <td>190080</td>\n"," <td>Arbre</td>\n"," <td>DASCO</td>\n"," <td>PARIS 20E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PLAINE (9) ELEM B</td>\n"," <td>11110006</td>\n"," <td>Tilleul</td>\n"," <td>Tilia</td>\n"," <td>tomentosa</td>\n"," <td>NaN</td>\n"," <td>18589</td>\n"," <td>15.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.40</td>\n"," </tr>\n"," <tr>\n"," <th>82951</th>\n"," <td>203295</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 18E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>RUE D AUBERVILLIERS</td>\n"," <td>301004</td>\n"," <td>Erable</td>\n"," <td>Acer</td>\n"," <td>monspessulanum</td>\n"," <td>NaN</td>\n"," <td>5545</td>\n"," <td>0.56</td>\n"," <td>J</td>\n"," <td>0.00</td>\n"," <td>48.89</td>\n"," <td>2.37</td>\n"," </tr>\n"," <tr>\n"," <th>90605</th>\n"," <td>211362</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 5E ARRDT</td>\n"," <td>34</td>\n"," <td>NaN</td>\n"," <td>BOULEVARD SAINT MARCEL</td>\n"," <td>204025</td>\n"," <td>Cedrele</td>\n"," <td>Toona</td>\n"," <td>sinensis</td>\n"," <td>NaN</td>\n"," <td>20130</td>\n"," <td>16.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.84</td>\n"," <td>2.36</td>\n"," </tr>\n"," <tr>\n"," <th>91256</th>\n"," <td>212024</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 12E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>RUE DE CHALON</td>\n"," <td>202005</td>\n"," <td>Micocoulier</td>\n"," <td>Celtis</td>\n"," <td>australis</td>\n"," <td>NaN</td>\n"," <td>30140</td>\n"," <td>10.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.37</td>\n"," </tr>\n"," <tr>\n"," <th>97354</th>\n"," <td>218274</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>27</td>\n"," <td>NaN</td>\n"," <td>AVENUE RAPP</td>\n"," <td>501003</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>occidentalis</td>\n"," <td>NaN</td>\n"," <td>8110</td>\n"," <td>18.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.30</td>\n"," </tr>\n"," <tr>\n"," <th>103437</th>\n"," <td>224609</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 18E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>BOULEVARD BARBES</td>\n"," <td>1303003</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>9035</td>\n"," <td>6.00</td>\n"," <td>J</td>\n"," <td>0.00</td>\n"," <td>48.89</td>\n"," <td>2.35</td>\n"," </tr>\n"," <tr>\n"," <th>133546</th>\n"," <td>256868</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 19E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>AVENUE JEAN JAURES</td>\n"," <td>1501006</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>4090</td>\n"," <td>10.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.89</td>\n"," <td>2.39</td>\n"," </tr>\n"," <tr>\n"," <th>147235</th>\n"," <td>270997</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>QUAI BRANLY</td>\n"," <td>503005</td>\n"," <td>Paulownia</td>\n"," <td>Paulownia</td>\n"," <td>tomentosa</td>\n"," <td>NaN</td>\n"," <td>9155</td>\n"," <td>15.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.29</td>\n"," </tr>\n"," <tr>\n"," <th>147634</th>\n"," <td>271402</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>AVENUE BOSQUET</td>\n"," <td>1202016</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>occidentalis</td>\n"," <td>NaN</td>\n"," <td>65110</td>\n"," <td>16.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.31</td>\n"," </tr>\n"," <tr>\n"," <th>148106</th>\n"," <td>271884</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>AVENUE BOSQUET</td>\n"," <td>501022</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>occidentalis</td>\n"," <td>NaN</td>\n"," <td>6595</td>\n"," <td>15.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.30</td>\n"," </tr>\n"," <tr>\n"," <th>148481</th>\n"," <td>272263</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>14</td>\n"," <td>NaN</td>\n"," <td>BOULEVARD DES INVALIDES</td>\n"," <td>402009</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>occidentalis</td>\n"," <td>NaN</td>\n"," <td>6120</td>\n"," <td>18.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.31</td>\n"," </tr>\n"," <tr>\n"," <th>149056</th>\n"," <td>272885</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 18E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PLACE DE LA CHAPELLE</td>\n"," <td>101026</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>250255</td>\n"," <td>22.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.88</td>\n"," <td>2.36</td>\n"," </tr>\n"," <tr>\n"," <th>151364</th>\n"," <td>275451</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ESPLANADE DES INVALIDES</td>\n"," <td>410005</td>\n"," <td>Tilleul</td>\n"," <td>Tilia</td>\n"," <td>tomentosa</td>\n"," <td>NaN</td>\n"," <td>7540</td>\n"," <td>8.15</td>\n"," <td>JA</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.31</td>\n"," </tr>\n"," <tr>\n"," <th>152303</th>\n"," <td>276396</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 7E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ESPLANADE DES INVALIDES</td>\n"," <td>402011</td>\n"," <td>Tilleul</td>\n"," <td>Tilia</td>\n"," <td>tomentosa</td>\n"," <td>NaN</td>\n"," <td>9085</td>\n"," <td>12.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.86</td>\n"," <td>2.31</td>\n"," </tr>\n"," <tr>\n"," <th>160432</th>\n"," <td>284823</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 18E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>AVENUE DE LA PORTE DE CLIGNANCOURT</td>\n"," <td>202005</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>orientalis</td>\n"," <td>NaN</td>\n"," <td>120170</td>\n"," <td>20.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.90</td>\n"," <td>2.34</td>\n"," </tr>\n"," <tr>\n"," <th>163918</th>\n"," <td>288407</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 12E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>BOULEVARD DE PICPUS</td>\n"," <td>1604017</td>\n"," <td>Sophora</td>\n"," <td>Sophora</td>\n"," <td>japonica</td>\n"," <td>NaN</td>\n"," <td>6085</td>\n"," <td>5.00</td>\n"," <td>JA</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.40</td>\n"," </tr>\n"," <tr>\n"," <th>164616</th>\n"," <td>289121</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 12E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>RUE FRANCOIS TRUFFAUT</td>\n"," <td>301005</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>9095</td>\n"," <td>10.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.83</td>\n"," <td>2.39</td>\n"," </tr>\n"," <tr>\n"," <th>167606</th>\n"," <td>292180</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 18E ARRDT</td>\n"," <td>16</td>\n"," <td>NaN</td>\n"," <td>RUE JEAN COCTEAU</td>\n"," <td>202027</td>\n"," <td>Noisetier de Byzance</td>\n"," <td>Corylus</td>\n"," <td>colurna</td>\n"," <td>NaN</td>\n"," <td>3630</td>\n"," <td>5.00</td>\n"," <td>J</td>\n"," <td>0.00</td>\n"," <td>48.90</td>\n"," <td>2.35</td>\n"," </tr>\n"," <tr>\n"," <th>170367</th>\n"," <td>295047</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 5E ARRDT</td>\n"," <td>16</td>\n"," <td>NaN</td>\n"," <td>RUE POLIVEAU</td>\n"," <td>202015</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>6125</td>\n"," <td>20.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.84</td>\n"," <td>2.36</td>\n"," </tr>\n"," <tr>\n"," <th>173343</th>\n"," <td>298211</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 5E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>BOULEVARD DE PORT ROYAL</td>\n"," <td>402001</td>\n"," <td>Marronnier</td>\n"," <td>Aesculus</td>\n"," <td>x carnea</td>\n"," <td>Briotii'</td>\n"," <td>79120</td>\n"," <td>10.00</td>\n"," <td>A</td>\n"," <td>0.00</td>\n"," <td>48.84</td>\n"," <td>2.35</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5076a33a-6a75-4619-8a9b-6c434233f810')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-5076a33a-6a75-4619-8a9b-6c434233f810 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-5076a33a-6a75-4619-8a9b-6c434233f810');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-ef18fe46-73ec-41f2-832b-2509630c4668\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-ef18fe46-73ec-41f2-832b-2509630c4668')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-ef18fe46-73ec-41f2-832b-2509630c4668 button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_ba4709f5-9b65-4e01-a39c-095ad73c8538\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('max_girth_outlier_df')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_ba4709f5-9b65-4e01-a39c-095ad73c8538 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('max_girth_outlier_df');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"max_girth_outlier_df","summary":"{\n \"name\": \"max_girth_outlier_df\",\n \"rows\": 20,\n \"fields\": [\n {\n \"column\": \"id\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 34759,\n \"min\": 190080,\n \"max\": 298211,\n \"num_unique_values\": 20,\n \"samples\": [\n 190080,\n 292180,\n 288407\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"type_emplacement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Arbre\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"domanialite\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2,\n \"samples\": [\n \"Alignement\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"arrondissement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 6,\n \"samples\": [\n \"PARIS 20E ARRDT\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"complement_addresse\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"27\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": null,\n \"max\": null,\n \"num_unique_values\": 0,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lieu\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 18,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"id_emplacement\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 19,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 9,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 9,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 10,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"circonference_cm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 59471,\n \"min\": 3630,\n \"max\": 250255,\n \"num_unique_values\": 20,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"hauteur_m\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 5.839249658898595,\n \"min\": 0.56,\n \"max\": 22.0,\n \"num_unique_values\": 11,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 3,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_a\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.020935631218655087,\n \"min\": 48.8345576057,\n \"max\": 48.8996027298,\n \"num_unique_values\": 20,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_b\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.03448434237606588,\n \"min\": 2.292727369,\n \"max\": 2.4008649434,\n \"num_unique_values\": 20,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":29}],"source":["print('girth outliers trees ('+ str(len(max_girth_outlier_df)) + ' occurrencies):')\n","max_girth_outlier_df"]},{"cell_type":"code","execution_count":30,"id":"3535207c-ddca-4791-afc8-924bdf424cbb","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":742},"id":"3535207c-ddca-4791-afc8-924bdf424cbb","executionInfo":{"status":"ok","timestamp":1737062566943,"user_tz":-60,"elapsed":56,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"e770ae74-760e-4527-97b9-3c3d361e3182"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["circonference_cm\n","3630 1\n","4090 1\n","120170 1\n","79120 1\n","65110 1\n","30140 1\n","20130 1\n","18589 1\n","9155 1\n","9095 1\n","9085 1\n","9035 1\n","8110 1\n","7540 1\n","6595 1\n","6125 1\n","6120 1\n","6085 1\n","5545 1\n","250255 1\n","Name: count, dtype: int64"],"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>count</th>\n"," </tr>\n"," <tr>\n"," <th>circonference_cm</th>\n"," <th></th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>3630</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>4090</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>120170</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>79120</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>65110</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>30140</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>20130</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>18589</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>9155</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>9095</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>9085</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>9035</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>8110</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>7540</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>6595</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>6125</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>6120</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>6085</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>5545</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>250255</th>\n"," <td>1</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div><br><label><b>dtype:</b> int64</label>"]},"metadata":{},"execution_count":30}],"source":["max_girth_outlier_df.value_counts('circonference_cm')"]},{"cell_type":"code","execution_count":31,"id":"e20288eb-4c78-4b56-b411-5d0c74599659","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":460},"id":"e20288eb-4c78-4b56-b411-5d0c74599659","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062566944,"user_tz":-60,"elapsed":54,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"4fa4d9d1-1291-45aa-ef2e-747de62a21fc"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["hauteur_m\n","10.00 4\n","15.00 3\n","5.00 2\n","16.00 2\n","18.00 2\n","20.00 2\n","0.56 1\n","6.00 1\n","8.15 1\n","12.00 1\n","22.00 1\n","Name: count, dtype: int64"],"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>count</th>\n"," </tr>\n"," <tr>\n"," <th>hauteur_m</th>\n"," <th></th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>10.00</th>\n"," <td>4</td>\n"," </tr>\n"," <tr>\n"," <th>15.00</th>\n"," <td>3</td>\n"," </tr>\n"," <tr>\n"," <th>5.00</th>\n"," <td>2</td>\n"," </tr>\n"," <tr>\n"," <th>16.00</th>\n"," <td>2</td>\n"," </tr>\n"," <tr>\n"," <th>18.00</th>\n"," <td>2</td>\n"," </tr>\n"," <tr>\n"," <th>20.00</th>\n"," <td>2</td>\n"," </tr>\n"," <tr>\n"," <th>0.56</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>6.00</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>8.15</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>12.00</th>\n"," <td>1</td>\n"," </tr>\n"," <tr>\n"," <th>22.00</th>\n"," <td>1</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div><br><label><b>dtype:</b> int64</label>"]},"metadata":{},"execution_count":31}],"source":["max_girth_outlier_df.value_counts('hauteur_m')"]},{"cell_type":"code","execution_count":32,"id":"Lw7drXq5HvUS","metadata":{"id":"Lw7drXq5HvUS","executionInfo":{"status":"ok","timestamp":1737062566944,"user_tz":-60,"elapsed":52,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# Set circonference_cm values to NaN for these indexes\n","df_refined.loc[max_girth_outlier_df.index, 'circonference_cm'] = np.nan"]},{"cell_type":"code","execution_count":33,"id":"7_jcJ0ufG4Qk","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"7_jcJ0ufG4Qk","executionInfo":{"status":"ok","timestamp":1737062566944,"user_tz":-60,"elapsed":52,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"e7a03437-b2af-4015-d591-cd0a7546005e"},"outputs":[{"output_type":"stream","name":"stdout","text":["Amount of to high girths : 20.\n","All values have been set to \"NaN\" and the cases will be reported\n"]}],"source":["print(f'Amount of to high girths : {len(max_girth_outlier_df)}.\\nAll values have been set to \"NaN\" and the cases will be reported')"]},{"cell_type":"markdown","id":"747b4e39-2678-40e9-af04-2e1caa765226","metadata":{"id":"747b4e39-2678-40e9-af04-2e1caa765226"},"source":["##### Min"]},{"cell_type":"markdown","id":"8f65de59-ef68-43b1-be1b-9c3ddada8372","metadata":{"id":"8f65de59-ef68-43b1-be1b-9c3ddada8372"},"source":["Height :"]},{"cell_type":"code","execution_count":34,"id":"6a24f15e-6a9e-46d0-acc1-fb7bf281ec30","metadata":{"id":"6a24f15e-6a9e-46d0-acc1-fb7bf281ec30","executionInfo":{"status":"ok","timestamp":1737062566944,"user_tz":-60,"elapsed":47,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# based on the tallest tree height : https://en.wikipedia.org/wiki/Hyperion_(tree)\n","min_height = 0 # in meters"]},{"cell_type":"code","execution_count":35,"id":"10c2ce83-27b5-43f1-af4f-5d80ac2ba48c","metadata":{"id":"10c2ce83-27b5-43f1-af4f-5d80ac2ba48c","executionInfo":{"status":"ok","timestamp":1737062566944,"user_tz":-60,"elapsed":46,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["min_height_outlier_df = df_refined[(df_refined['hauteur_m'] <= min_height)]"]},{"cell_type":"code","execution_count":36,"id":"4dfbde45-9d17-4a5f-af04-282461462235","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":146},"id":"4dfbde45-9d17-4a5f-af04-282461462235","executionInfo":{"status":"ok","timestamp":1737062566944,"user_tz":-60,"elapsed":46,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"bc42741e-ad98-4124-d280-a926786bdf5a"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["hauteur_m\n","0.00 39219\n","Name: count, dtype: int64"],"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>count</th>\n"," </tr>\n"," <tr>\n"," <th>hauteur_m</th>\n"," <th></th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0.00</th>\n"," <td>39219</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div><br><label><b>dtype:</b> int64</label>"]},"metadata":{},"execution_count":36}],"source":["# no negative value : ok\n","min_height_outlier_df.value_counts('hauteur_m')"]},{"cell_type":"code","execution_count":37,"id":"fa7fd3b4-7e48-4289-b005-939f565b889e","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":999},"id":"fa7fd3b4-7e48-4289-b005-939f565b889e","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062566944,"user_tz":-60,"elapsed":44,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"b6e95c76-ebcb-4a76-952f-4cdeb933d2bf"},"outputs":[{"output_type":"stream","name":"stdout","text":["height outliers trees (39219 occurrencies):\n"]},{"output_type":"execute_result","data":{"text/plain":[" id type_emplacement domanialite arrondissement \\\n","4 99878 Arbre Jardin PARIS 17E ARRDT \n","5 99879 Arbre Jardin PARIS 17E ARRDT \n","6 99880 Arbre Jardin PARIS 17E ARRDT \n","11 99885 Arbre Jardin PARIS 16E ARRDT \n","12 99887 Arbre Jardin PARIS 16E ARRDT \n","... ... ... ... ... \n","200132 2024741 Arbre Jardin BOIS DE VINCENNES \n","200133 2024742 Arbre Jardin BOIS DE VINCENNES \n","200134 2024743 Arbre Jardin BOIS DE VINCENNES \n","200135 2024744 Arbre Jardin BOIS DE VINCENNES \n","200136 2024745 Arbre Jardin BOIS DE VINCENNES \n","\n"," complement_addresse numero \\\n","4 NaN NaN \n","5 NaN NaN \n","6 NaN NaN \n","11 NaN NaN \n","12 NaN NaN \n","... ... ... \n","200132 NaN NaN \n","200133 NaN NaN \n","200134 NaN NaN \n","200135 NaN NaN \n","200136 NaN NaN \n","\n"," lieu id_emplacement \\\n","4 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000G0037 \n","5 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000G0036 \n","6 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000G0035 \n","11 JARDIN DU RANELAGH 20001 \n","12 JARDIN DU RANELAGH 20003 \n","... ... ... \n","200132 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170151 \n","200133 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170152 \n","200134 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170153 \n","200135 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170154 \n","200136 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170155 \n","\n"," libelle_francais genre espece variete \\\n","4 Arbre à miel Tetradium daniellii NaN \n","5 Arbre à miel Tetradium daniellii NaN \n","6 Arbre à miel Tetradium daniellii NaN \n","11 Hêtre Fagus sylvatica Atropunicea' \n","12 Micocoulier Celtis occidentalis NaN \n","... ... ... ... ... \n","200132 NaN Castanea fargesii NaN \n","200133 Arbre aux mouchoirs Davidia involucrata NaN \n","200134 Peuplier Populus simonii NaN \n","200135 Chêne Quercus n. sp. NaN \n","200136 Raisinier Hovenia dulcis NaN \n","\n"," circonference_cm hauteur_m stade_developpement remarquable \\\n","4 38.00 0.00 NaN NaN \n","5 38.00 0.00 NaN NaN \n","6 37.00 0.00 NaN NaN \n","11 30.00 0.00 NaN NaN \n","12 205.00 0.00 NaN NaN \n","... ... ... ... ... \n","200132 0.00 0.00 NaN 0.00 \n","200133 0.00 0.00 NaN 0.00 \n","200134 0.00 0.00 NaN 0.00 \n","200135 0.00 0.00 NaN 0.00 \n","200136 0.00 0.00 NaN 0.00 \n","\n"," geo_point_2d_a geo_point_2d_b \n","4 48.89 2.32 \n","5 48.89 2.32 \n","6 48.89 2.32 \n","11 48.86 2.27 \n","12 48.86 2.27 \n","... ... ... \n","200132 48.82 2.46 \n","200133 48.82 2.46 \n","200134 48.82 2.46 \n","200135 48.82 2.46 \n","200136 48.82 2.45 \n","\n","[39219 rows x 18 columns]"],"text/html":["\n"," <div id=\"df-a418bd76-0698-4fe9-8885-88962425efdc\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>type_emplacement</th>\n"," <th>domanialite</th>\n"," <th>arrondissement</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>lieu</th>\n"," <th>id_emplacement</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>4</th>\n"," <td>99878</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000G0037</td>\n"," <td>Arbre à miel</td>\n"," <td>Tetradium</td>\n"," <td>daniellii</td>\n"," <td>NaN</td>\n"," <td>38.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>5</th>\n"," <td>99879</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000G0036</td>\n"," <td>Arbre à miel</td>\n"," <td>Tetradium</td>\n"," <td>daniellii</td>\n"," <td>NaN</td>\n"," <td>38.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>6</th>\n"," <td>99880</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000G0035</td>\n"," <td>Arbre à miel</td>\n"," <td>Tetradium</td>\n"," <td>daniellii</td>\n"," <td>NaN</td>\n"," <td>37.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>11</th>\n"," <td>99885</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 16E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>JARDIN DU RANELAGH</td>\n"," <td>20001</td>\n"," <td>Hêtre</td>\n"," <td>Fagus</td>\n"," <td>sylvatica</td>\n"," <td>Atropunicea'</td>\n"," <td>30.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.86</td>\n"," <td>2.27</td>\n"," </tr>\n"," <tr>\n"," <th>12</th>\n"," <td>99887</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 16E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>JARDIN DU RANELAGH</td>\n"," <td>20003</td>\n"," <td>Micocoulier</td>\n"," <td>Celtis</td>\n"," <td>occidentalis</td>\n"," <td>NaN</td>\n"," <td>205.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.86</td>\n"," <td>2.27</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"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>200132</th>\n"," <td>2024741</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170151</td>\n"," <td>NaN</td>\n"," <td>Castanea</td>\n"," <td>fargesii</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200133</th>\n"," <td>2024742</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170152</td>\n"," <td>Arbre aux mouchoirs</td>\n"," <td>Davidia</td>\n"," <td>involucrata</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200134</th>\n"," <td>2024743</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170153</td>\n"," <td>Peuplier</td>\n"," <td>Populus</td>\n"," <td>simonii</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200135</th>\n"," <td>2024744</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170154</td>\n"," <td>Chêne</td>\n"," <td>Quercus</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200136</th>\n"," <td>2024745</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170155</td>\n"," <td>Raisinier</td>\n"," <td>Hovenia</td>\n"," <td>dulcis</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.45</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>39219 rows × 18 columns</p>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a418bd76-0698-4fe9-8885-88962425efdc')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-a418bd76-0698-4fe9-8885-88962425efdc button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-a418bd76-0698-4fe9-8885-88962425efdc');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-f200ca0f-d525-4107-a124-3fe1cbcb536a\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-f200ca0f-d525-4107-a124-3fe1cbcb536a')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-f200ca0f-d525-4107-a124-3fe1cbcb536a button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_48d7496d-6030-4f12-b43d-24ff307e3712\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('min_height_outlier_df')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_48d7496d-6030-4f12-b43d-24ff307e3712 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('min_height_outlier_df');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"min_height_outlier_df","summary":"{\n \"name\": \"min_height_outlier_df\",\n \"rows\": 39219,\n \"fields\": [\n {\n \"column\": \"id\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 635080,\n \"min\": 99878,\n \"max\": 2024745,\n \"num_unique_values\": 39219,\n \"samples\": [\n 123467,\n 2011241,\n 2020241\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"type_emplacement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Arbre\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"domanialite\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 7,\n \"samples\": [\n \"Jardin\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"arrondissement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 25,\n \"samples\": [\n \"PARIS 13E ARRDT\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"complement_addresse\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 120,\n \"samples\": [\n \"15 (XVI-7603)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": null,\n \"max\": null,\n \"num_unique_values\": 0,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lieu\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2397,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"id_emplacement\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 24787,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 170,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 156,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 462,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 256,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"circonference_cm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 51.57964421106181,\n \"min\": 0.0,\n \"max\": 900.0,\n \"num_unique_values\": 298,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"hauteur_m\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 4,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_a\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.04342193848100632,\n \"min\": 48.754504992,\n \"max\": 48.9106150304,\n \"num_unique_values\": 39213,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_b\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0577776801177361,\n \"min\": 2.2102412288,\n \"max\": 2.4621650013,\n \"num_unique_values\": 39214,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":37}],"source":["# Maybe freshly planted trees are 0 meters tall and/or 0 is a default value when not measured\n","print('height outliers trees ('+ str(len(min_height_outlier_df)) + ' occurrencies):')\n","min_height_outlier_df"]},{"cell_type":"code","execution_count":38,"id":"ad996a3c-d4b9-4ff6-8e39-eca8054157ff","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"ad996a3c-d4b9-4ff6-8e39-eca8054157ff","executionInfo":{"status":"ok","timestamp":1737062566945,"user_tz":-60,"elapsed":38,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"68397898-f18b-4d89-c5bc-af2677f5491e"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" stade_developpement counts\n","0 NaN 36952\n","1 A 1191\n","2 JA 577\n","3 J 474\n","4 M 25"],"text/html":["\n"," <div id=\"df-ae5253ef-96af-415b-8504-cda6c03e1625\" class=\"colab-df-container\">\n"," <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>stade_developpement</th>\n"," <th>counts</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>NaN</td>\n"," <td>36952</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>A</td>\n"," <td>1191</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>JA</td>\n"," <td>577</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>J</td>\n"," <td>474</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>M</td>\n"," <td>25</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ae5253ef-96af-415b-8504-cda6c03e1625')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-ae5253ef-96af-415b-8504-cda6c03e1625 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-ae5253ef-96af-415b-8504-cda6c03e1625');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-7b06270f-2d30-4822-81b9-6e52c7abed54\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-7b06270f-2d30-4822-81b9-6e52c7abed54')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-7b06270f-2d30-4822-81b9-6e52c7abed54 button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_c98d4c9b-13bc-4667-96b6-710be0ae5ec9\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('min_height_value_counts')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_c98d4c9b-13bc-4667-96b6-710be0ae5ec9 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('min_height_value_counts');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"min_height_value_counts","summary":"{\n \"name\": \"min_height_value_counts\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"JA\",\n \"M\",\n \"A\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"counts\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 16277,\n \"min\": 25,\n \"max\": 36952,\n \"num_unique_values\": 5,\n \"samples\": [\n 1191,\n 25,\n 577\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":38}],"source":["# A = Adult, JA = Jeune Adulte, J = Jeune, M = Mature\n","# Majority has no stade_developpement value. Probably freshly planted grain. Recommend to not ignore these trees.\n","# For the others, it may probably be a fill value. I recommend setting the height values as NaN\n","min_height_value_counts = min_height_outlier_df.value_counts('stade_developpement', dropna=False).reset_index(name='counts')\n","min_height_value_counts"]},{"cell_type":"code","execution_count":39,"id":"34d32b28-1a5f-4df0-a620-4c418befd07c","metadata":{"id":"34d32b28-1a5f-4df0-a620-4c418befd07c","executionInfo":{"status":"ok","timestamp":1737062566945,"user_tz":-60,"elapsed":36,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# giving NaN value instead of 0 for already existing trees in order to signal missing data\n","zero_height_but_existing_development_state = (df_refined['hauteur_m'] <= min_height) & (df_refined['stade_developpement'].isin(['A', 'JA', 'J', 'M']))\n","# display(df_refined[zero_girth_but_existing_development_state].head(1).transpose())\n","df_refined.loc[zero_height_but_existing_development_state,'hauteur_m'] = np.nan\n","\n","# giving NaN value instead of 0 for already existing trees in order to signal missing data\n","# df_refined.loc[(df_refined['hauteur_m'] <= min_height) & (df_refined['stade_developpement'].isin(['A', 'JA', 'J', 'M'])),\n","# 'hauteur_m'] = np.nan\n","# check 1 or 2 trees tagged as 'A' or 'JA' (or other) to see if these really is a positive height"]},{"cell_type":"code","execution_count":40,"id":"zILrPvh-JjKp","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"zILrPvh-JjKp","executionInfo":{"status":"ok","timestamp":1737062566945,"user_tz":-60,"elapsed":35,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"5cba9d21-3806-483e-ae80-f5e1555c5bb4"},"outputs":[{"output_type":"stream","name":"stdout","text":["2267 trees have a zero height value whereas they belong to one of these categories : ['A', 'JA', 'J', 'M'] --> height value set to 'NaN'\n","36952 trees have a zero height with no 'stade_developpement' value. They probably are freshly planted grains : height value is kept zero\n"]}],"source":["print(f\"{min_height_value_counts['counts'].sum()-min_height_value_counts['counts'][0]} trees have a zero height value whereas they belong to one of these categories : ['A', 'JA', 'J', 'M'] --> height value set to 'NaN'\")\n","print(f\"{min_height_value_counts['counts'][0]} trees have a zero height with no 'stade_developpement' value. They probably are freshly planted grains : height value is kept zero\")\n"]},{"cell_type":"markdown","id":"27fbf69c-619d-40ee-8514-2b3c3470007b","metadata":{"id":"27fbf69c-619d-40ee-8514-2b3c3470007b"},"source":["Girth :"]},{"cell_type":"code","execution_count":41,"id":"0d1a2d77-c7d8-4944-9a58-dc045519c731","metadata":{"id":"0d1a2d77-c7d8-4944-9a58-dc045519c731","executionInfo":{"status":"ok","timestamp":1737062566945,"user_tz":-60,"elapsed":31,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# based on the biggest tree girth (circumference) : https://www.guinnessworldrecords.com/world-records/living-tree-with-greatest-girth\n","min_girth = 0 # in meters"]},{"cell_type":"code","execution_count":42,"id":"fe7e6808-fa76-4e9c-bddf-27474aa9f28c","metadata":{"id":"fe7e6808-fa76-4e9c-bddf-27474aa9f28c","executionInfo":{"status":"ok","timestamp":1737062566945,"user_tz":-60,"elapsed":30,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["min_girth_outlier_df = df_refined[df_refined['circonference_cm'] <= min_girth]"]},{"cell_type":"code","execution_count":43,"id":"6baa2bd7-0136-451c-aa81-4f61c30ba125","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":146},"id":"6baa2bd7-0136-451c-aa81-4f61c30ba125","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062566946,"user_tz":-60,"elapsed":30,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"db6ed9e6-7bb2-41a2-ef32-47adf07fa2d0"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["circonference_cm\n","0.00 25867\n","Name: count, dtype: int64"],"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>count</th>\n"," </tr>\n"," <tr>\n"," <th>circonference_cm</th>\n"," <th></th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0.00</th>\n"," <td>25867</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div><br><label><b>dtype:</b> int64</label>"]},"metadata":{},"execution_count":43}],"source":["# no negative value : ok\n","min_girth_outlier_df.value_counts('circonference_cm')"]},{"cell_type":"code","execution_count":44,"id":"9f015caa-3e56-437c-9e6d-e315754d33a7","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"id":"9f015caa-3e56-437c-9e6d-e315754d33a7","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062567527,"user_tz":-60,"elapsed":610,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"e26ad911-abc4-4fd8-99be-c9db6f8d3617"},"outputs":[{"output_type":"stream","name":"stdout","text":["girth outliers trees (25867 occurrencies):\n"]},{"output_type":"execute_result","data":{"text/plain":[" id type_emplacement domanialite arrondissement \\\n","1778 101922 Arbre Jardin PARIS 19E ARRDT \n","2332 102591 Arbre Jardin PARIS 17E ARRDT \n","2333 102592 Arbre Jardin PARIS 17E ARRDT \n","2405 102744 Arbre Jardin PARIS 20E ARRDT \n","2430 102773 Arbre Jardin PARIS 17E ARRDT \n","... ... ... ... ... \n","200132 2024741 Arbre Jardin BOIS DE VINCENNES \n","200133 2024742 Arbre Jardin BOIS DE VINCENNES \n","200134 2024743 Arbre Jardin BOIS DE VINCENNES \n","200135 2024744 Arbre Jardin BOIS DE VINCENNES \n","200136 2024745 Arbre Jardin BOIS DE VINCENNES \n","\n"," complement_addresse numero \\\n","1778 NaN NaN \n","2332 NaN NaN \n","2333 NaN NaN \n","2405 NaN NaN \n","2430 NaN NaN \n","... ... ... \n","200132 NaN NaN \n","200133 NaN NaN \n","200134 NaN NaN \n","200135 NaN NaN \n","200136 NaN NaN \n","\n"," lieu id_emplacement \\\n","1778 PARC DES BUTTES CHAUMONT C0400032 \n","2332 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000B0052 \n","2333 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000B0053 \n","2405 SQUARE LEON FRAPIE 2 \n","2430 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000B0021 \n","... ... ... \n","200132 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170151 \n","200133 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170152 \n","200134 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170153 \n","200135 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170154 \n","200136 ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F... 720170155 \n","\n"," libelle_francais genre espece variete \\\n","1778 Erable Acer platanoides NaN \n","2332 Magnolia Magnolia x soulangeana Heaven Scent' \n","2333 Magnolia Magnolia x soulangeana Heaven Scent' \n","2405 Platane Platanus x hispanica NaN \n","2430 Magnolia Magnolia x soulangeana Sundew' \n","... ... ... ... ... \n","200132 NaN Castanea fargesii NaN \n","200133 Arbre aux mouchoirs Davidia involucrata NaN \n","200134 Peuplier Populus simonii NaN \n","200135 Chêne Quercus n. sp. NaN \n","200136 Raisinier Hovenia dulcis NaN \n","\n"," circonference_cm hauteur_m stade_developpement remarquable \\\n","1778 0.00 0.00 NaN NaN \n","2332 0.00 0.00 NaN NaN \n","2333 0.00 0.00 NaN NaN \n","2405 0.00 NaN JA 0.00 \n","2430 0.00 0.00 NaN NaN \n","... ... ... ... ... \n","200132 0.00 0.00 NaN 0.00 \n","200133 0.00 0.00 NaN 0.00 \n","200134 0.00 0.00 NaN 0.00 \n","200135 0.00 0.00 NaN 0.00 \n","200136 0.00 0.00 NaN 0.00 \n","\n"," geo_point_2d_a geo_point_2d_b \n","1778 48.88 2.38 \n","2332 48.89 2.32 \n","2333 48.89 2.32 \n","2405 48.87 2.41 \n","2430 48.89 2.32 \n","... ... ... \n","200132 48.82 2.46 \n","200133 48.82 2.46 \n","200134 48.82 2.46 \n","200135 48.82 2.46 \n","200136 48.82 2.45 \n","\n","[25867 rows x 18 columns]"],"text/html":["\n"," <div id=\"df-cbf945f5-6991-4b82-b326-3ad3fdbd846d\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>type_emplacement</th>\n"," <th>domanialite</th>\n"," <th>arrondissement</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>lieu</th>\n"," <th>id_emplacement</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>1778</th>\n"," <td>101922</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 19E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC DES BUTTES CHAUMONT</td>\n"," <td>C0400032</td>\n"," <td>Erable</td>\n"," <td>Acer</td>\n"," <td>platanoides</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.88</td>\n"," <td>2.38</td>\n"," </tr>\n"," <tr>\n"," <th>2332</th>\n"," <td>102591</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000B0052</td>\n"," <td>Magnolia</td>\n"," <td>Magnolia</td>\n"," <td>x soulangeana</td>\n"," <td>Heaven Scent'</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>2333</th>\n"," <td>102592</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000B0053</td>\n"," <td>Magnolia</td>\n"," <td>Magnolia</td>\n"," <td>x soulangeana</td>\n"," <td>Heaven Scent'</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>2405</th>\n"," <td>102744</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 20E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>SQUARE LEON FRAPIE</td>\n"," <td>2</td>\n"," <td>Platane</td>\n"," <td>Platanus</td>\n"," <td>x hispanica</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>JA</td>\n"," <td>0.00</td>\n"," <td>48.87</td>\n"," <td>2.41</td>\n"," </tr>\n"," <tr>\n"," <th>2430</th>\n"," <td>102773</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000B0021</td>\n"," <td>Magnolia</td>\n"," <td>Magnolia</td>\n"," <td>x soulangeana</td>\n"," <td>Sundew'</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</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"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>200132</th>\n"," <td>2024741</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170151</td>\n"," <td>NaN</td>\n"," <td>Castanea</td>\n"," <td>fargesii</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200133</th>\n"," <td>2024742</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170152</td>\n"," <td>Arbre aux mouchoirs</td>\n"," <td>Davidia</td>\n"," <td>involucrata</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200134</th>\n"," <td>2024743</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170153</td>\n"," <td>Peuplier</td>\n"," <td>Populus</td>\n"," <td>simonii</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200135</th>\n"," <td>2024744</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170154</td>\n"," <td>Chêne</td>\n"," <td>Quercus</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.46</td>\n"," </tr>\n"," <tr>\n"," <th>200136</th>\n"," <td>2024745</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>BOIS DE VINCENNES</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>ARBORETUM DE L ECOLE DU BREUIL - ROUTE DE LA F...</td>\n"," <td>720170155</td>\n"," <td>Raisinier</td>\n"," <td>Hovenia</td>\n"," <td>dulcis</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.82</td>\n"," <td>2.45</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>25867 rows × 18 columns</p>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cbf945f5-6991-4b82-b326-3ad3fdbd846d')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-cbf945f5-6991-4b82-b326-3ad3fdbd846d button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-cbf945f5-6991-4b82-b326-3ad3fdbd846d');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-9277bca7-97b4-4a15-ab5c-c0b3ed11370f\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-9277bca7-97b4-4a15-ab5c-c0b3ed11370f')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-9277bca7-97b4-4a15-ab5c-c0b3ed11370f button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_acfeb848-64c5-4969-9482-aedd9ddfe56c\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('min_girth_outlier_df')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_acfeb848-64c5-4969-9482-aedd9ddfe56c button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('min_girth_outlier_df');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"min_girth_outlier_df","summary":"{\n \"name\": \"min_girth_outlier_df\",\n \"rows\": 25867,\n \"fields\": [\n {\n \"column\": \"id\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 721969,\n \"min\": 101922,\n \"max\": 2024745,\n \"num_unique_values\": 25867,\n \"samples\": [\n 2016263,\n 145520,\n 169124\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"type_emplacement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Arbre\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"domanialite\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 7,\n \"samples\": [\n \"Jardin\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"arrondissement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 24,\n \"samples\": [\n \"PARIS 12E ARRDT\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"complement_addresse\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 60,\n \"samples\": [\n \"2 (face)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": null,\n \"max\": null,\n \"num_unique_values\": 0,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lieu\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2101,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"id_emplacement\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 21389,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 152,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 141,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 412,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 134,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"circonference_cm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"hauteur_m\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.9520728384383992,\n \"min\": 0.0,\n \"max\": 40.0,\n \"num_unique_values\": 20,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 4,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_a\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.04898659494827352,\n \"min\": 48.754504992,\n \"max\": 48.9104979364,\n \"num_unique_values\": 25862,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_b\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.058212997253484625,\n \"min\": 2.2260475851,\n \"max\": 2.4621650013,\n \"num_unique_values\": 25863,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":44}],"source":["# Maybe freshly planted trees have 0 cms girth and/or 0 is a default value when not measured\n","print('girth outliers trees ('+ str(len(min_girth_outlier_df)) + ' occurrencies):')\n","min_girth_outlier_df"]},{"cell_type":"code","execution_count":45,"id":"6cb2665b-7f8f-4004-b68a-449a6965b461","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":206},"id":"6cb2665b-7f8f-4004-b68a-449a6965b461","executionInfo":{"status":"ok","timestamp":1737062567527,"user_tz":-60,"elapsed":22,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"1d78791d-098a-46f1-a168-a5accee3e653"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" stade_developpement counts\n","0 NaN 25586\n","1 J 126\n","2 A 81\n","3 JA 51\n","4 M 23"],"text/html":["\n"," <div id=\"df-cc7fe3ce-fa1c-4ba6-8461-a17ef5d1deb8\" class=\"colab-df-container\">\n"," <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>stade_developpement</th>\n"," <th>counts</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>NaN</td>\n"," <td>25586</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>J</td>\n"," <td>126</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>A</td>\n"," <td>81</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>JA</td>\n"," <td>51</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>M</td>\n"," <td>23</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-cc7fe3ce-fa1c-4ba6-8461-a17ef5d1deb8')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-cc7fe3ce-fa1c-4ba6-8461-a17ef5d1deb8 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-cc7fe3ce-fa1c-4ba6-8461-a17ef5d1deb8');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-2e7d563c-61e0-4468-ada6-c23c33cd0150\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-2e7d563c-61e0-4468-ada6-c23c33cd0150')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-2e7d563c-61e0-4468-ada6-c23c33cd0150 button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_429c7c64-380f-4e43-b261-ffe7e4c9b717\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('min_girth_value_counts')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_429c7c64-380f-4e43-b261-ffe7e4c9b717 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('min_girth_value_counts');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"min_girth_value_counts","summary":"{\n \"name\": \"min_girth_value_counts\",\n \"rows\": 5,\n \"fields\": [\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 4,\n \"samples\": [\n \"A\",\n \"M\",\n \"J\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"counts\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 11411,\n \"min\": 23,\n \"max\": 25586,\n \"num_unique_values\": 5,\n \"samples\": [\n 126,\n 23,\n 81\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":45}],"source":["# A = Adult, JA = Jeune Adulte, J = Jeune, M = Mature\n","# Majority has no stade_developpement value. Probably freshly planted grain. Recommend to not ignore these trees.\n","# For the others, it may probably be a fill value. I recommend setting the height values as NaN\n","min_girth_value_counts = min_girth_outlier_df.value_counts('stade_developpement', dropna=False).reset_index(name='counts')\n","min_girth_value_counts"]},{"cell_type":"code","execution_count":46,"id":"bf4cf441-c5af-4ebc-9ab7-92dfaeefe10c","metadata":{"id":"bf4cf441-c5af-4ebc-9ab7-92dfaeefe10c","executionInfo":{"status":"ok","timestamp":1737062567528,"user_tz":-60,"elapsed":22,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# giving NaN value instead of 0 for already existing trees in order to signal missing data\n","zero_girth_but_existing_development_state = (df_refined['circonference_cm'] <= min_girth) & (df_refined['stade_developpement'].isin(['A', 'JA', 'J', 'M']))\n","# display(df_refined[zero_girth_but_existing_development_state].head(1).transpose())\n","df_refined.loc[zero_girth_but_existing_development_state,'circonference_cm'] = np.nan\n","# same question as minimum height (see above)"]},{"cell_type":"code","execution_count":47,"id":"BLeJcjd0MJub","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"BLeJcjd0MJub","executionInfo":{"status":"ok","timestamp":1737062567528,"user_tz":-60,"elapsed":21,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"35ae275f-671c-4718-d138-954a5212a3d9"},"outputs":[{"output_type":"stream","name":"stdout","text":["281 trees have a zero girth value whereas they belong to one of these categories : ['A', 'JA', 'J', 'M'] --> girth value set to 'NaN'\n","25586 trees have a zero girth with no 'stade_developpement' value. They probably are freshly planted grains : girth value is kept zero\n"]}],"source":["print(f\"{min_girth_value_counts['counts'].sum()-min_girth_value_counts['counts'][0]} trees have a zero girth value whereas they belong to one of these categories : ['A', 'JA', 'J', 'M'] --> girth value set to 'NaN'\")\n","print(f\"{min_girth_value_counts['counts'][0]} trees have a zero girth with no 'stade_developpement' value. They probably are freshly planted grains : girth value is kept zero\")\n"]},{"cell_type":"markdown","id":"1MPpoe9IOLP1","metadata":{"id":"1MPpoe9IOLP1"},"source":["##### Girth and Height coherence"]},{"cell_type":"code","execution_count":48,"id":"e3e28272-33a0-428a-80e8-975140d30330","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"e3e28272-33a0-428a-80e8-975140d30330","executionInfo":{"status":"ok","timestamp":1737062567528,"user_tz":-60,"elapsed":18,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"1cb1e637-0963-4a69-dfd2-7b00754616a4"},"outputs":[{"output_type":"stream","name":"stdout","text":["25264 trees are both 0 m tall and 0 cms girth --> they probably are freshly planted grains\n","11688 trees are 0 m tall but have strictly positive and coherent girth --> height values are set to 'NaN' \n","322 trees have strictly positive and coherent height but 0 cm girth --> girth values are set to 'NaN' \n"]}],"source":["from IPython.display import display\n","\n","# search if trees 0 meters height trees and 0 cms girth trees are the same\n","both_zero = ((df_refined['hauteur_m'] <= 0) & (df_refined['circonference_cm'] <= 0)).sum()\n","print(f\"{both_zero} trees are both 0 m tall and 0 cms girth --> they probably are freshly planted grains\")\n","\n","height_zero_girth_non_zero = (df_refined['hauteur_m'] <= 0) & (df_refined['circonference_cm'] > 0)\n","# display(df_refined[height_zero_girth_non_zero].head(1).transpose())\n","df_refined.loc[height_zero_girth_non_zero,'hauteur_m'] = np.nan\n","print(f\"{height_zero_girth_non_zero.sum()} trees are 0 m tall but have strictly positive and coherent girth --> height values are set to 'NaN' \")\n","\n","non_zero_height_zero_girth = (df_refined['hauteur_m'] > 0) & (df_refined['circonference_cm'] <= 0)\n","# display(df_refined[non_zero_height_zero_girth].head(1).transpose())\n","df_refined.loc[non_zero_height_zero_girth,'circonference_cm'] = np.nan\n","print(f\"{non_zero_height_zero_girth.sum()} trees have strictly positive and coherent height but 0 cm girth --> girth values are set to 'NaN' \")\n","\n"]},{"cell_type":"markdown","id":"26f2d2ae-4489-4d6e-b217-16125f372597","metadata":{"id":"26f2d2ae-4489-4d6e-b217-16125f372597"},"source":["### NaN Values"]},{"cell_type":"markdown","id":"bf1ee819-3acf-420e-8d17-3699bbc1b4ac","metadata":{"id":"bf1ee819-3acf-420e-8d17-3699bbc1b4ac"},"source":["```\n","# domanialite can be found when looking for nearest neighbors\n","# complement_adresse and numero often do not exist\n","# Maybe can we do something about the missing values for \"libelle_francais\", \"genre\", \"espece\", \"variete\"\n","# stade_developpement and remarkable missing values presences seem to be correlated (according to previous analysis)\n","```\n","\n","\n"]},{"cell_type":"code","execution_count":49,"id":"01c33327-96f8-4cca-bd48-af40c023344d","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":89},"id":"01c33327-96f8-4cca-bd48-af40c023344d","scrolled":true,"executionInfo":{"status":"ok","timestamp":1737062567528,"user_tz":-60,"elapsed":15,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"b9b94480-90c3-4b3a-c383-ffe4f7c35c1e"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" domanialite complement_addresse numero libelle_francais genre espece \\\n","0 1 169235 200137 1497 16 1752 \n","\n"," variete stade_developpement remarquable \n","0 163360 67205 63098 "],"text/html":["\n"," <div id=\"df-5e0e7113-5f93-4e2c-add8-a102587e4e04\" class=\"colab-df-container\">\n"," <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>domanialite</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>1</td>\n"," <td>169235</td>\n"," <td>200137</td>\n"," <td>1497</td>\n"," <td>16</td>\n"," <td>1752</td>\n"," <td>163360</td>\n"," <td>67205</td>\n"," <td>63098</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-5e0e7113-5f93-4e2c-add8-a102587e4e04')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-5e0e7113-5f93-4e2c-add8-a102587e4e04 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-5e0e7113-5f93-4e2c-add8-a102587e4e04');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n"," <div id=\"id_a19e2d2e-6609-44e8-83e3-11494ae45440\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('columns_with_nans_df')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_a19e2d2e-6609-44e8-83e3-11494ae45440 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('columns_with_nans_df');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"columns_with_nans_df","summary":"{\n \"name\": \"columns_with_nans_df\",\n \"rows\": 1,\n \"fields\": [\n {\n \"column\": \"domanialite\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 1,\n \"max\": 1,\n \"num_unique_values\": 1,\n \"samples\": [\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"complement_addresse\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 169235,\n \"max\": 169235,\n \"num_unique_values\": 1,\n \"samples\": [\n 169235\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 200137,\n \"max\": 200137,\n \"num_unique_values\": 1,\n \"samples\": [\n 200137\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 1497,\n \"max\": 1497,\n \"num_unique_values\": 1,\n \"samples\": [\n 1497\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 16,\n \"max\": 16,\n \"num_unique_values\": 1,\n \"samples\": [\n 16\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 1752,\n \"max\": 1752,\n \"num_unique_values\": 1,\n \"samples\": [\n 1752\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 163360,\n \"max\": 163360,\n \"num_unique_values\": 1,\n \"samples\": [\n 163360\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 67205,\n \"max\": 67205,\n \"num_unique_values\": 1,\n \"samples\": [\n 67205\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 63098,\n \"max\": 63098,\n \"num_unique_values\": 1,\n \"samples\": [\n 63098\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":49}],"source":["columns_with_nans_df"]},{"cell_type":"markdown","id":"hBvGSxoNq7H2","metadata":{"id":"hBvGSxoNq7H2"},"source":["#### domanialite"]},{"cell_type":"code","execution_count":50,"id":"z_8W9BlA4wnW","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"z_8W9BlA4wnW","executionInfo":{"status":"ok","timestamp":1737062567528,"user_tz":-60,"elapsed":14,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"a0e0a3ba-f162-4b32-e0e0-2eba5d095e85"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[48.85761995, 2.32096211],\n"," [48.85765607, 2.32103142],\n"," [48.85770501, 2.32106057],\n"," ...,\n"," [48.82290714, 2.45797007],\n"," [48.82252234, 2.45595555],\n"," [48.82044479, 2.45485622]])"]},"metadata":{},"execution_count":50}],"source":["df_refined[['geo_point_2d_a','geo_point_2d_b']].values"]},{"cell_type":"code","execution_count":51,"id":"D27qtLjtqEnA","metadata":{"id":"D27qtLjtqEnA","executionInfo":{"status":"ok","timestamp":1737062567528,"user_tz":-60,"elapsed":12,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["from data_analysis_functions import compute_distances"]},{"cell_type":"code","execution_count":52,"id":"cIuY0Unh2g7r","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"cIuY0Unh2g7r","executionInfo":{"status":"ok","timestamp":1737062567528,"user_tz":-60,"elapsed":12,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"70ddfa96-8143-4ab4-cb95-dc8cd3c57f14"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([[48.84954716, 2.41418967]])"]},"metadata":{},"execution_count":52}],"source":["nan_indexes = df_refined[df_refined['domanialite'].isna()].index\n","nan_indexes\n","df_refined.loc[nan_indexes, ['geo_point_2d_a','geo_point_2d_b']].values"]},{"cell_type":"code","execution_count":53,"id":"70FaYnAPq1N4","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":71946,"status":"ok","timestamp":1737062639465,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"70FaYnAPq1N4","outputId":"5683878c-4a12-4c1d-9c6b-db1cc21be630"},"outputs":[{"output_type":"stream","name":"stdout","text":["Mean distance between tree 197239 and its 3 nearest neighbours: 6.02 meters\n","Missing values for domanialite: Jardin\n"]}],"source":["for index in nan_indexes:\n"," distances = compute_distances(df_refined, index)\n"," # find three nearest neighbours\n"," three_nearest_neighbours = distances.nsmallest(4)[1:]\n"," print(f'Mean distance between tree {index} and its 3 nearest neighbours: {three_nearest_neighbours.mean().round(2)} meters')\n"," missing_domanialite = df_refined.loc[three_nearest_neighbours.index,'domanialite'].unique()\n"," if len(missing_domanialite) == 1:\n"," print(f'Missing values for domanialite: {missing_domanialite[0]}')\n"," df_refined.loc[index, 'domanialite'] = missing_domanialite[0]\n"," else:\n"," print(f'Too many values in missing_domanialite: {missing_domanialite}')"]},{"cell_type":"markdown","id":"cD1g4VFyJgH6","metadata":{"id":"cD1g4VFyJgH6"},"source":["#### genre"]},{"cell_type":"code","execution_count":54,"id":"balAs4z2Je3F","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":550},"executionInfo":{"elapsed":10,"status":"ok","timestamp":1737062639465,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"balAs4z2Je3F","outputId":"db78d6c9-f750-44fb-869f-447f3880be21"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" genre espece variete libelle_francais\n","195404 NaN NaN NaN NaN\n","195409 NaN NaN NaN NaN\n","195410 NaN NaN NaN NaN\n","195475 NaN NaN NaN NaN\n","195476 NaN NaN NaN NaN\n","195487 NaN NaN NaN NaN\n","195496 NaN NaN NaN NaN\n","195497 NaN NaN NaN NaN\n","195499 NaN NaN NaN NaN\n","195502 NaN NaN NaN NaN\n","195503 NaN NaN NaN NaN\n","195505 NaN NaN NaN NaN\n","195509 NaN NaN NaN NaN\n","195514 NaN NaN NaN NaN\n","195518 NaN NaN NaN NaN\n","195519 NaN NaN NaN NaN"],"text/html":["\n"," <div id=\"df-642244c9-c71e-4cde-81d0-3473c01591a6\" class=\"colab-df-container\">\n"," <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>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>libelle_francais</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>195404</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195409</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195410</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195475</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195476</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195487</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195496</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195497</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195499</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195502</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195503</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195505</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195509</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195514</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195518</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>195519</th>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-642244c9-c71e-4cde-81d0-3473c01591a6')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-642244c9-c71e-4cde-81d0-3473c01591a6 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-642244c9-c71e-4cde-81d0-3473c01591a6');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-7851b789-fed2-42f4-bbe4-002d70a4e3e3\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-7851b789-fed2-42f4-bbe4-002d70a4e3e3')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-7851b789-fed2-42f4-bbe4-002d70a4e3e3 button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_46a0504c-961d-480b-9736-b89cf69ce2d0\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('genre_nan_rows')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_46a0504c-961d-480b-9736-b89cf69ce2d0 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('genre_nan_rows');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"genre_nan_rows","repr_error":"0"}},"metadata":{},"execution_count":54}],"source":["# catch each line where \"genre\" value is nan. Catch only 'genre','espece','variete','libelle_francais' columns\n","\n","# Assuming df_refined is your DataFrame as defined in the provided code.\n","\n","genre_nan_rows = df_refined[df_refined['genre'].isna()][['genre', 'espece', 'variete', 'libelle_francais']]\n","\n","genre_nan_rows"]},{"cell_type":"markdown","id":"MgBXUNswKdtn","metadata":{"id":"MgBXUNswKdtn"},"source":["Cannot find \"genre\" value since all the other values for \"espece\", \"variete\", \"libelle_francais\" are \"nan\"."]},{"cell_type":"markdown","id":"jPyVZ2cQLLJ0","metadata":{"id":"jPyVZ2cQLLJ0"},"source":["#### libelle_francais"]},{"cell_type":"code","execution_count":55,"id":"ACltGPqrSWo0","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":457,"status":"ok","timestamp":1737062639914,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"ACltGPqrSWo0","outputId":"a8729550-dbac-4366-cc1b-f46bdbdfa335"},"outputs":[{"output_type":"stream","name":"stdout","text":["Number of different libelle_francais values: 193\n","True\n"]}],"source":["print(f\"Number of different libelle_francais values: {len(np.unique(df_refined['libelle_francais'].astype(str)))}\")\n","# nan is in the list. We will continue ignoring it in order to potentially find a correspondance between libelle_francais and genre variable\n","print('nan' in np.unique(df_refined['libelle_francais'].astype(str)))"]},{"cell_type":"code","execution_count":56,"id":"GbXIdFi6T5rR","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":6,"status":"ok","timestamp":1737062639914,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"GbXIdFi6T5rR","outputId":"ec1153cd-8df5-48ab-c581-b2655cf31652"},"outputs":[{"output_type":"stream","name":"stdout","text":["Number of different libelle_francais values: 192\n","Possible correlation between libelle_francais and genre:\n","Number of 'libelle_francais' with one 'genre': 182\n","Number of 'libelle_francais' with multiple 'genre': 10\n"]}],"source":["# not the most useful\n","# For each value of \"libelle_francais\", find when there is only one genre possible.\n","# Count the cases when there is only one genre related to libelle_francais and the cases where it is not.\n","# Please consider the \"nan\" cases in another counter variable\n","\n","# Count different 'genre' values for each 'libelle_francais'. NaNs are ignored\n","libelle_genre_counts = df_refined.groupby('libelle_francais')['genre'].nunique()\n","\n","one_genre_count = (libelle_genre_counts == 1).sum()\n","multiple_genres_count = (libelle_genre_counts > 1).sum()\n","\n","nan_genre_count = df_refined[(df_refined['libelle_francais'].notna()) & (df_refined['genre'].isna())].shape[0]\n","\n","print(f\"Number of different libelle_francais values: {len(libelle_genre_counts)}\")\n","print('Possible correlation between libelle_francais and genre:')\n","print(f\"Number of 'libelle_francais' with one 'genre': {one_genre_count}\")\n","print(f\"Number of 'libelle_francais' with multiple 'genre': {multiple_genres_count}\")"]},{"cell_type":"code","execution_count":57,"id":"pEffiwi8gfAH","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":5,"status":"ok","timestamp":1737062639914,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"pEffiwi8gfAH","outputId":"15d85929-e170-46eb-c1db-77c934963961"},"outputs":[{"output_type":"stream","name":"stdout","text":["Number of different genre values: 175\n","Possible correlation between genre and libelle_francais:\n","Number of 'genre' with one libelle_francais: 159\n","Number of 'genre' with multiple libelle_francais: 13\n","Number of 'genre' with no libelle_francais: 3\n","\n","Genre that did not match any libelle_francais value: genre\n","Crataegomespilus 0\n","Podocarpus 0\n","x Sorbopyrus 0\n","Name: libelle_francais, dtype: int64\n"]}],"source":["# For each value of \"genre\", find when there is only one libelle_francais possible.\n","# Count the cases when there is only one libelle_francais related to genre and the cases where it is not.\n","# Please consider the \"nan\" cases in another counter variable\n","\n","# Count different 'libelle_francais' values for each 'genre'. NaNs are ignored\n","genre_libelle_counts = df_refined.groupby('genre')['libelle_francais'].nunique()\n","\n","one_libelle_count = (genre_libelle_counts == 1).sum()\n","multiple_libelles_count = (genre_libelle_counts > 1).sum()\n","no_libelles_count = (genre_libelle_counts < 1).sum()\n","\n","print(f\"Number of different genre values: {len(genre_libelle_counts)}\")\n","print('Possible correlation between genre and libelle_francais:')\n","print(f\"Number of 'genre' with one libelle_francais: {one_libelle_count}\")\n","print(f\"Number of 'genre' with multiple libelle_francais: {multiple_libelles_count}\")\n","print(f\"Number of 'genre' with no libelle_francais: {no_libelles_count}\")\n","print(f\"\\nGenre that did not match any libelle_francais value: {genre_libelle_counts[genre_libelle_counts < 1]}\")\n","\n"]},{"cell_type":"code","execution_count":58,"id":"dVhtzF2YeW-_","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":3289,"status":"ok","timestamp":1737062643201,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"dVhtzF2YeW-_","outputId":"99292d34-7fd8-41c0-b0dc-878084eedc2a"},"outputs":[{"output_type":"stream","name":"stdout","text":["One libelle mapping dictionary: (159 values)\n","\n","More than one libelle mapping dictionary: (13 values)\n","\n","No libelle mapping dictionary: (3 values)\n","{'Crataegomespilus': None, 'Podocarpus': None, 'x Sorbopyrus': None}\n"]}],"source":["# Generate a dictionary that contains matching libelle_francais for each genre, based on the df_refined dataframe\n","\n","# Create a dictionary that maps each 'genre' to its corresponding 'libelle_francais'\n","# Only include 'genre' that have a unique 'libelle_francais'\n","only_one_libelle_mapping = {}\n","more_than_one_libelle_mapping = {}\n","no_libelle_mapping = {}\n","\n","for genre in df_refined['genre'].dropna().unique(): # Iterating over 'genre'\n"," libelles = df_refined[df_refined['genre'] == genre]['libelle_francais'].dropna().unique() # Getting 'libelle_francais'\n"," if len(libelles) == 1:\n"," only_one_libelle_mapping[genre] = libelles[0] # Mapping genre to single libelle_francais\n"," elif len(libelles) > 1:\n"," more_than_one_libelle_mapping[genre] = libelles # Mapping genre to multiple libelle_francais\n"," else:\n"," no_libelle_mapping[genre] = None # Mapping genre to no libelle_francais\n","\n","\n","from pprint import pprint\n","print(f\"One libelle mapping dictionary: ({len(only_one_libelle_mapping)} values)\")\n","# pprint(only_one_libelle_mapping)\n","print()\n","print(f\"More than one libelle mapping dictionary: ({len(more_than_one_libelle_mapping)} values)\")\n","# pprint(more_than_one_libelle_mapping)\n","print()\n","print(f\"No libelle mapping dictionary: ({len(no_libelle_mapping)} values)\")\n","pprint(no_libelle_mapping)"]},{"cell_type":"code","execution_count":59,"id":"9F7qfwNmjsl-","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":607},"executionInfo":{"elapsed":14,"status":"ok","timestamp":1737062643201,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"9F7qfwNmjsl-","outputId":"7b859b05-f52b-4359-cae8-be9fb41bbfa3"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x600 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["# prompt: plot bars with len of each dictionnary\n","\n","import matplotlib.pyplot as plt\n","\n","# Assuming you have the dictionaries: only_one_libelle_mapping, more_than_one_libelle_mapping, no_libelle_mapping\n","\n","dictionaries = {\n"," \"= 1\": only_one_libelle_mapping,\n"," \"> 1\": more_than_one_libelle_mapping,\n"," \"= 0\": no_libelle_mapping,\n","}\n","\n","dictionary_names = list(dictionaries.keys())\n","lengths = [len(dictionaries[name]) for name in dictionary_names]\n","\n","plt.figure(figsize=(10, 6))\n","plt.bar(dictionary_names, lengths)\n","plt.xlabel(\"Distinct libelle_francais corresponding values per genre\")\n","plt.ylabel(\"Ocurrences\")\n","plt.tight_layout()\n","plt.show()"]},{"cell_type":"code","execution_count":60,"id":"XFSIHEGQjZfa","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":13,"status":"ok","timestamp":1737062643201,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"XFSIHEGQjZfa","outputId":"500f1c3c-dc45-42ba-bd51-8963a897772d"},"outputs":[{"output_type":"stream","name":"stdout","text":["{'x Sorbopyrus': None, 'Podocarpus': None, 'Crataegomespilus': None}\n"," genre espece variete libelle_francais\n","199193 x Sorbopyrus n. sp. NaN NaN\n","199527 Podocarpus macrophyllus NaN NaN\n","199528 Podocarpus macrophyllus NaN NaN\n","200040 Crataegomespilus grandiflora NaN NaN\n","200049 Crataegomespilus grandiflora NaN NaN\n","\"x Sorbopyrus\":\"Poirier de Bollwiller\" entry correctly added to the dictionary\n","\"Podocarpus\":\"Pin des Bouddhistes\" entry correctly added to the dictionary\n","\"Crataegomespilus\":\"Neflier à grandes feuilles\" entry correctly added to the dictionary\n"]}],"source":["# Case \"=0\" corresponding libelle_francais values per genre : add the information\n","print(no_libelle_mapping)\n","print(df_refined.loc[df_refined['genre'].isin(no_libelle_mapping.keys()),['genre','espece','variete','libelle_francais']])\n","# https://lapepinieredufruitier.fr/categorie-produit/poirier/sorbopyrus/\n","# https://fr.wikipedia.org/wiki/Podocarpus_macrophyllus\n","# https://www.jardin-ecologique.fr/boutique/toutes-nos-plantes/plantes-mediterraneennes/plantes-pour-haies/haie-gourmande/crataegomespilus-grandifolia-neflier-a-grandes-feuilles/#:~:text=Le%20Crataegomespilus%20est%20un%20hybride,ou%20aussi%20x%20Crataegomespilus%20grandiflora\n","no_libelle_mapping['x Sorbopyrus'] = 'Poirier de Bollwiller'\n","no_libelle_mapping['Podocarpus'] = 'Pin des Bouddhistes'\n","no_libelle_mapping['Crataegomespilus'] = 'Neflier à grandes feuilles'\n","\n","for index, row in no_libelle_mapping.items():\n"," if index not in only_one_libelle_mapping.keys():\n"," only_one_libelle_mapping[index] = row\n"," print(f'\"{index}\":\"{row}\" entry correctly added to the dictionary')\n"," else:\n"," print(f'Problem with \"{index}\" : found in both dictionaries')"]},{"cell_type":"code","execution_count":61,"id":"Bm65nzDZnJuv","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":35},"executionInfo":{"elapsed":10,"status":"ok","timestamp":1737062643202,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"Bm65nzDZnJuv","outputId":"76b60506-de08-46d4-cc96-8508f938b6e0"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["'Poirier de Bollwiller'"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"string"}},"metadata":{},"execution_count":61}],"source":["only_one_libelle_mapping['x Sorbopyrus']"]},{"cell_type":"code","execution_count":62,"id":"Gia4bWbb3vWr","metadata":{"executionInfo":{"elapsed":13638,"status":"ok","timestamp":1737062656831,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"Gia4bWbb3vWr"},"outputs":[],"source":["# for each line in df_refined where libelle_francais is missing (nan value),\n","# if genre value is not nan and genre value is in only_one_libelle_mapping dictionary, use the corresponding value in only_one_libelle_mapping as libelle_francais value\n","\n","for index, row in df_refined.iterrows():\n"," if pd.isna(row['libelle_francais']):\n"," if not pd.isna(row['genre']) and row['genre'] in only_one_libelle_mapping:\n"," df_refined.loc[index, 'libelle_francais'] = only_one_libelle_mapping[row['genre']]"]},{"cell_type":"code","execution_count":63,"id":"UuUT5S3m44Bp","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":15,"status":"ok","timestamp":1737062656831,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"UuUT5S3m44Bp","outputId":"f1e16e49-4eb7-48b2-d3f4-566a1c24dc22"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["array([ 0, 0, 0, 0, 169235, 200137, 0, 0,\n"," 304, 16, 1752, 163360, 623, 13956, 67205, 63098,\n"," 0, 0])"]},"metadata":{},"execution_count":63}],"source":["# number of 'nan' for each column\n","column_nb_of_nans = np.array([], dtype='int')\n","for column in df_refined.columns:\n"," where_nan = df_refined[df_refined[column].isna()]\n"," column_nb_of_nans = np.concatenate((column_nb_of_nans,[len(where_nan)]))\n","column_nb_of_nans"]},{"cell_type":"code","execution_count":64,"id":"fSQZsKAo44Q_","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":107},"executionInfo":{"elapsed":13,"status":"ok","timestamp":1737062656831,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"fSQZsKAo44Q_","outputId":"758a959e-3bef-47e6-f7bf-e77391b71371","scrolled":true},"outputs":[{"output_type":"stream","name":"stdout","text":["Number of nan for each variable that is concerned by 1 nan or more:\n"]},{"output_type":"execute_result","data":{"text/plain":[" complement_addresse numero libelle_francais genre espece variete \\\n","0 169235 200137 304 16 1752 163360 \n","\n"," circonference_cm hauteur_m stade_developpement remarquable \n","0 623 13956 67205 63098 "],"text/html":["\n"," <div id=\"df-e06234b6-3de7-4ee5-887b-c06ca0645ec0\" class=\"colab-df-container\">\n"," <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>complement_addresse</th>\n"," <th>numero</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>169235</td>\n"," <td>200137</td>\n"," <td>304</td>\n"," <td>16</td>\n"," <td>1752</td>\n"," <td>163360</td>\n"," <td>623</td>\n"," <td>13956</td>\n"," <td>67205</td>\n"," <td>63098</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e06234b6-3de7-4ee5-887b-c06ca0645ec0')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-e06234b6-3de7-4ee5-887b-c06ca0645ec0 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-e06234b6-3de7-4ee5-887b-c06ca0645ec0');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n"," <div id=\"id_3c0b4c17-3c80-4020-a292-eeb95ceeae22\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('columns_with_nans_df')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_3c0b4c17-3c80-4020-a292-eeb95ceeae22 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('columns_with_nans_df');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"columns_with_nans_df","summary":"{\n \"name\": \"columns_with_nans_df\",\n \"rows\": 1,\n \"fields\": [\n {\n \"column\": \"complement_addresse\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 169235,\n \"max\": 169235,\n \"num_unique_values\": 1,\n \"samples\": [\n 169235\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 200137,\n \"max\": 200137,\n \"num_unique_values\": 1,\n \"samples\": [\n 200137\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 304,\n \"max\": 304,\n \"num_unique_values\": 1,\n \"samples\": [\n 304\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 16,\n \"max\": 16,\n \"num_unique_values\": 1,\n \"samples\": [\n 16\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 1752,\n \"max\": 1752,\n \"num_unique_values\": 1,\n \"samples\": [\n 1752\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 163360,\n \"max\": 163360,\n \"num_unique_values\": 1,\n \"samples\": [\n 163360\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"circonference_cm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 623,\n \"max\": 623,\n \"num_unique_values\": 1,\n \"samples\": [\n 623\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"hauteur_m\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 13956,\n \"max\": 13956,\n \"num_unique_values\": 1,\n \"samples\": [\n 13956\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"stade_developpement\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 67205,\n \"max\": 67205,\n \"num_unique_values\": 1,\n \"samples\": [\n 67205\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 63098,\n \"max\": 63098,\n \"num_unique_values\": 1,\n \"samples\": [\n 63098\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":64}],"source":["# dataframe with number of nans per column only when one nan or more\n","where_nans = np.where(column_nb_of_nans>0)\n","columns_with_nans_df = pd.DataFrame([column_nb_of_nans[where_nans]], columns = df_refined.columns[where_nans])\n","print('Number of nan for each variable that is concerned by 1 nan or more:')\n","columns_with_nans_df"]},{"cell_type":"code","execution_count":65,"id":"Vpj0tzTt1zBZ","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":9,"status":"ok","timestamp":1737062656831,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"Vpj0tzTt1zBZ","outputId":"f459d844-f18e-46a6-ea5b-e6a0f4d464be"},"outputs":[{"output_type":"execute_result","data":{"text/plain":["{'Prunus': array(['Prunus n. sp.', 'Cerisier à fleurs', 'Prunier à fleurs',\n"," 'Prunier à fruits', 'Pêcher', 'Merisier', 'Laurier-Cerise',\n"," 'Laurier du Portugal', 'Cerisier à grappes', 'Cerisier à fruits',\n"," 'Amandier', 'Prunellier', 'Laurier du caucase',\n"," 'Prunus Fleur n. sp.', 'Abricotier', 'Prunus Fruit n. sp.',\n"," 'Laurier des Açores'], dtype=object),\n"," 'Ligustrum': array(['Troene', 'Troëne'], dtype=object),\n"," 'Ostrya': array(['Charme-Houblon', 'Ostryer'], dtype=object),\n"," 'Malus': array(['Pommier à fleurs', 'Pommier', 'Pommier à fruits'], dtype=object),\n"," 'Pyrus': array(['Poirier', 'Poirier à fleurs', 'Poirier à fruits'], dtype=object),\n"," 'Fraxinus': array(['Frêne', 'Frêne à fleurs'], dtype=object),\n"," 'Crataegus': array(['Aubépine', 'Aubepine'], dtype=object),\n"," 'Corylus': array(['Noisetier de Byzance', 'Noisetier'], dtype=object),\n"," 'Sorbus': array(['Alisier', 'Sorbier', 'Cormier'], dtype=object),\n"," 'Parrotia': array(['Parrotie de Perse - Arbre de fer', 'Parrotie de Perse'],\n"," dtype=object),\n"," 'Gymnocladus': array(['Chicot du Canada', 'Fevier'], dtype=object),\n"," 'Laurus': array(['Laurier sauce', 'Laurier'], dtype=object),\n"," 'Cotinus': array(['Arbre à perruque', \"Fustet d' Amérique\"], dtype=object)}"]},"metadata":{},"execution_count":65}],"source":["# Case \">1 \" corresponding libelle_francais values per genre : check espece and (eventually) variete\n","more_than_one_libelle_mapping"]},{"cell_type":"code","execution_count":66,"id":"I2lF8TSH2TDZ","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":423},"executionInfo":{"elapsed":2582,"status":"ok","timestamp":1737062659407,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"I2lF8TSH2TDZ","outputId":"292720b7-73e5-4045-d955-da482dea3dd7"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" libelle_francais genre espece variete\n","0 Prunus n. sp. Prunus n. sp. NaN\n","1 Cerisier à fleurs Prunus serrulata NaN\n","2 Cerisier à fleurs Prunus serrulata Kanzan'\n","3 Cerisier à fleurs Prunus serrulata Hizakura'\n","4 Cerisier à fleurs Prunus serrulata Fugenzo'\n",".. ... ... ... ...\n","285 Fevier Gymnocladus n. sp. NaN\n","286 Laurier sauce Laurus nobilis NaN\n","287 Laurier Laurus n. sp. NaN\n","288 Arbre à perruque Cotinus coggygria NaN\n","289 Fustet d' Amérique Cotinus obovatus NaN\n","\n","[290 rows x 4 columns]"],"text/html":["\n"," <div id=\"df-dd5bee98-15b8-4c62-a42e-e1d3e7998edf\" class=\"colab-df-container\">\n"," <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>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>0</th>\n"," <td>Prunus n. sp.</td>\n"," <td>Prunus</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>1</th>\n"," <td>Cerisier à fleurs</td>\n"," <td>Prunus</td>\n"," <td>serrulata</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>2</th>\n"," <td>Cerisier à fleurs</td>\n"," <td>Prunus</td>\n"," <td>serrulata</td>\n"," <td>Kanzan'</td>\n"," </tr>\n"," <tr>\n"," <th>3</th>\n"," <td>Cerisier à fleurs</td>\n"," <td>Prunus</td>\n"," <td>serrulata</td>\n"," <td>Hizakura'</td>\n"," </tr>\n"," <tr>\n"," <th>4</th>\n"," <td>Cerisier à fleurs</td>\n"," <td>Prunus</td>\n"," <td>serrulata</td>\n"," <td>Fugenzo'</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>285</th>\n"," <td>Fevier</td>\n"," <td>Gymnocladus</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>286</th>\n"," <td>Laurier sauce</td>\n"," <td>Laurus</td>\n"," <td>nobilis</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>287</th>\n"," <td>Laurier</td>\n"," <td>Laurus</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>288</th>\n"," <td>Arbre à perruque</td>\n"," <td>Cotinus</td>\n"," <td>coggygria</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>289</th>\n"," <td>Fustet d' Amérique</td>\n"," <td>Cotinus</td>\n"," <td>obovatus</td>\n"," <td>NaN</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>290 rows × 4 columns</p>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-dd5bee98-15b8-4c62-a42e-e1d3e7998edf')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-dd5bee98-15b8-4c62-a42e-e1d3e7998edf button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-dd5bee98-15b8-4c62-a42e-e1d3e7998edf');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-97ca770b-51b0-491c-a9b9-b503e6ab680c\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-97ca770b-51b0-491c-a9b9-b503e6ab680c')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-97ca770b-51b0-491c-a9b9-b503e6ab680c button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_89c926ac-326c-4a5b-8891-bb77e185b270\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('matching_libelle_francais')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_89c926ac-326c-4a5b-8891-bb77e185b270 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('matching_libelle_francais');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"matching_libelle_francais","summary":"{\n \"name\": \"matching_libelle_francais\",\n \"rows\": 290,\n \"fields\": [\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 44,\n \"samples\": [\n \"Parrotie de Perse\",\n \"Poirier\",\n \"Poirier \\u00e0 fleurs\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 13,\n \"samples\": [\n \"Laurus\",\n \"Parrotia\",\n \"Prunus\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 118,\n \"samples\": [\n \"prunifolia\",\n \"fecunda\",\n \"x subhirtella\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 154,\n \"samples\": [\n \"Plena'\",\n \"Reinette Blanche du Canada'\",\n \"Gibsii'\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":66}],"source":["# For each couple (genre, libelle_francais) with genre the key of more_than_one_libelle_mapping and libelle_francais taking\n","# its value in the different corresponding lists, find if there are only one or more corresponding couple \"espece\" value in df_refined that fits the couple.\n","# If there is exactly one possible \"espece\" value, fill a dataframe with libelle_francais, genre and espece values.\n","# A 4th column for \"variete\" contains a NaN value. If there is more than one possible \"espece\" value, then check the different values for \"variete\" and fill the dataframe\n","# with the different cases\n","\n","# Create an empty DataFrame to store the results\n","matching_libelle_francais = pd.DataFrame(columns=['libelle_francais', 'genre', 'espece', 'variete'])\n","\n","# Iterate through the 'more_than_one_libelle_mapping' dictionary\n","for genre, libelles_francais in more_than_one_libelle_mapping.items():\n"," for libelle_francais in libelles_francais:\n"," # Find corresponding rows in df_refined\n"," matching_rows = df_refined[(df_refined['genre'] == genre) & (df_refined['libelle_francais'] == libelle_francais)]\n","\n"," # Check the number of unique 'espece' values\n"," unique_especes = matching_rows['espece'].unique()\n"," num_unique_especes = len(unique_especes)\n","\n"," if num_unique_especes == 1:\n"," # Exactly one 'espece' value\n"," espece = unique_especes[0]\n"," new_row = pd.DataFrame({'libelle_francais': [libelle_francais],\n"," 'genre': [genre],\n"," 'espece': [espece],\n"," 'variete': [np.nan]})\n"," matching_libelle_francais = pd.concat([matching_libelle_francais, new_row], ignore_index=True)\n"," elif num_unique_especes > 1:\n"," # More than one 'espece' value, check 'variete'\n"," for espece in unique_especes:\n"," variete_values = matching_rows[matching_rows['espece'] == espece]['variete'].unique()\n"," for variete in variete_values:\n"," new_row = pd.DataFrame({'libelle_francais': [libelle_francais],\n"," 'genre': [genre],\n"," 'espece': [espece],\n"," 'variete': [variete]})\n"," matching_libelle_francais = pd.concat([matching_libelle_francais, new_row], ignore_index=True)\n","\n","# Display the resulting DataFrame\n","matching_libelle_francais"]},{"cell_type":"code","execution_count":67,"id":"x2cpyOBisR1S","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000},"executionInfo":{"elapsed":9,"status":"ok","timestamp":1737062659407,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"x2cpyOBisR1S","outputId":"b100b9b3-073c-425e-dbd6-6f2a43b89c0c"},"outputs":[{"output_type":"stream","name":"stdout","text":["id 100589\n","type_emplacement Arbre\n","domanialite Jardin\n","arrondissement PARIS 7E ARRDT\n","complement_addresse NaN\n","numero NaN\n","lieu JARDIN DU CHAMP DE MARS\n","id_emplacement P0120631\n","libelle_francais NaN\n","genre Crataegus\n","espece japonica\n","variete NaN\n","circonference_cm 43.00\n","hauteur_m 3.00\n","stade_developpement A\n","remarquable 0.00\n","geo_point_2d_a 48.86\n","geo_point_2d_b 2.30\n","Name: 528, dtype: object\n"]},{"output_type":"execute_result","data":{"text/plain":[" libelle_francais genre espece \\\n","225 Aubépine Crataegus laevigata \n","226 Aubépine Crataegus laevigata \n","227 Aubépine Crataegus laevigata \n","228 Aubépine Crataegus x media \n","229 Aubépine Crataegus x media \n","230 Aubépine Crataegus x media \n","231 Aubépine Crataegus x lavalleei \n","232 Aubépine Crataegus x lavalleei \n","233 Aubépine Crataegus crus-galli \n","234 Aubépine Crataegus crus-galli \n","235 Aubépine Crataegus monogyna var. monogyna \n","236 Aubépine Crataegus monogyna \n","237 Aubépine Crataegus viridis \n","238 Aubépine Crataegus chrysocarpa var. phoenicea \n","239 Aubépine Crataegus x lettermanii \n","240 Aubépine Crataegus x sorbifolia \n","241 Aubépine Crataegus x persimilis \n","242 Aubépine Crataegus fecunda \n","243 Aubépine Crataegus canbyi \n","244 Aubépine Crataegus spathulata \n","245 Aubépine Crataegus aprica \n","246 Aubépine Crataegus ambigua \n","247 Aubépine Crataegus venustula \n","248 Aubepine Crataegus n. sp. \n","249 Aubepine Crataegus monogyna \n","250 Aubepine Crataegus rotundifolia var. aboriginum \n","251 Aubepine Crataegus champlainensis \n","\n"," variete \n","225 NaN \n","226 Paul's Scarlet Hamlet' \n","227 Plena' \n","228 Rubra Plena' \n","229 Punicea' \n","230 Paul's Scarlet' \n","231 Carrierei' \n","232 NaN \n","233 Salicifolia' \n","234 NaN \n","235 NaN \n","236 Stricta' \n","237 NaN \n","238 NaN \n","239 NaN \n","240 NaN \n","241 NaN \n","242 NaN \n","243 NaN \n","244 NaN \n","245 NaN \n","246 NaN \n","247 NaN \n","248 NaN \n","249 NaN \n","250 NaN \n","251 NaN "],"text/html":["\n"," <div id=\"df-4d85d17e-f1ca-45aa-8e0b-a890326e9f20\" class=\"colab-df-container\">\n"," <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>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>225</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>laevigata</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>226</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>laevigata</td>\n"," <td>Paul's Scarlet Hamlet'</td>\n"," </tr>\n"," <tr>\n"," <th>227</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>laevigata</td>\n"," <td>Plena'</td>\n"," </tr>\n"," <tr>\n"," <th>228</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x media</td>\n"," <td>Rubra Plena'</td>\n"," </tr>\n"," <tr>\n"," <th>229</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x media</td>\n"," <td>Punicea'</td>\n"," </tr>\n"," <tr>\n"," <th>230</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x media</td>\n"," <td>Paul's Scarlet'</td>\n"," </tr>\n"," <tr>\n"," <th>231</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x lavalleei</td>\n"," <td>Carrierei'</td>\n"," </tr>\n"," <tr>\n"," <th>232</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x lavalleei</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>233</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>crus-galli</td>\n"," <td>Salicifolia'</td>\n"," </tr>\n"," <tr>\n"," <th>234</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>crus-galli</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>235</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>monogyna var. monogyna</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>236</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>monogyna</td>\n"," <td>Stricta'</td>\n"," </tr>\n"," <tr>\n"," <th>237</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>viridis</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>238</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>chrysocarpa var. phoenicea</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>239</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x lettermanii</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>240</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x sorbifolia</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>241</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>x persimilis</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>242</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>fecunda</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>243</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>canbyi</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>244</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>spathulata</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>245</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>aprica</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>246</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>ambigua</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>247</th>\n"," <td>Aubépine</td>\n"," <td>Crataegus</td>\n"," <td>venustula</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>248</th>\n"," <td>Aubepine</td>\n"," <td>Crataegus</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>249</th>\n"," <td>Aubepine</td>\n"," <td>Crataegus</td>\n"," <td>monogyna</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>250</th>\n"," <td>Aubepine</td>\n"," <td>Crataegus</td>\n"," <td>rotundifolia var. aboriginum</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>251</th>\n"," <td>Aubepine</td>\n"," <td>Crataegus</td>\n"," <td>champlainensis</td>\n"," <td>NaN</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-4d85d17e-f1ca-45aa-8e0b-a890326e9f20')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-4d85d17e-f1ca-45aa-8e0b-a890326e9f20 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-4d85d17e-f1ca-45aa-8e0b-a890326e9f20');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-f3112e7c-e56b-4bf2-9426-c0036018daed\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-f3112e7c-e56b-4bf2-9426-c0036018daed')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-f3112e7c-e56b-4bf2-9426-c0036018daed button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n \"name\": \" (matching_libelle_francais['genre'] == row['genre'])]\",\n \"rows\": 27,\n \"fields\": [\n {\n \"column\": \"libelle_francais\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2,\n \"samples\": [\n \"Aubepine\",\n \"Aub\\u00e9pine\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"genre\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Crataegus\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"espece\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 20,\n \"samples\": [\n \"laevigata\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"variete\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 8,\n \"samples\": [\n \"Plena'\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":67}],"source":["row = df_refined[df_refined['libelle_francais'].isna()].iloc[0]\n","print(row)\n","matching_libelle_francais[\n"," (matching_libelle_francais['genre'] == row['genre'])]"]},{"cell_type":"code","execution_count":68,"id":"14KyAGnNqLaD","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"collapsed":true,"executionInfo":{"elapsed":669,"status":"ok","timestamp":1737062660069,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"14KyAGnNqLaD","outputId":"9618ce17-a00c-4f52-84f1-159c0ab6afcd"},"outputs":[{"output_type":"stream","name":"stdout","text":["Amount of 'libelle_francais' found: 142\n","Amount of 'libelle_francais' not found: 162\n"]}],"source":["# in df_refined, for each row where a nan value is found at 'libelle_francais' column, check the different \"libelle_francais\" corresponding values in matching_libelle_francais\n","# that fit the couple (\"genre\",\"espece\"). If there is only one possible \"libelle_francais\" value, then attribute it to the considered row.\n","# If not, also check the \"variete\" value and find out if there only is one \"libelle_francais\" possible matching value. If still not, note this case as unsolvable.\n","\n","index_found = []\n","index_not_found = []\n","\n","# Iterate through rows with NaN 'libelle_francais' in df_refined\n","for index, row in df_refined[df_refined['libelle_francais'].isna()].iterrows():\n"," # Find matching rows in matching_libelle_francais based on 'genre' and 'espece'\n","\n"," matching_rows = matching_libelle_francais[\n"," (matching_libelle_francais['genre'].astype(str) == str(row['genre'])) &\n"," (matching_libelle_francais['espece'].astype(str) == str(row['espece']))\n"," ]\n","\n"," # Check the number of unique 'libelle_francais' values\n"," num_unique_libelles = len(matching_rows['libelle_francais'].unique())\n","\n"," if num_unique_libelles == 1:\n"," # Only one possible 'libelle_francais'\n"," df_refined.loc[index, 'libelle_francais'] = matching_rows['libelle_francais'].iloc[0]\n"," # print(f\"'libelle_francais' value found : for index {index}\")\n"," index_found.append(index)\n","\n"," elif num_unique_libelles > 1:\n"," # Multiple possible 'libelle_francais', check 'variete'\n"," matching_rows_variete = matching_rows[\n"," matching_rows['variete'].astype(str) == str(row['variete'])\n"," ]\n"," num_unique_libelles_variete = len(matching_rows_variete['libelle_francais'].unique())\n","\n"," if num_unique_libelles_variete == 1:\n"," # Only one possible 'libelle_francais' after considering 'variete'\n"," df_refined.loc[index, 'libelle_francais'] = matching_rows_variete['libelle_francais'].iloc[0]\n"," # print(f\"'libelle_francais' value found : for index {index}\")\n"," index_found.append(index)\n"," else: # > 1\n"," # Still multiple possible 'libelle_francais', mark as unsolvable\n"," # print(f\"'libelle_francais' value not found : for index {index} : too many possible values\")\n"," index_not_found.append(index)\n","\n"," else: # num_unique_libelles == 0\n"," # No possible 'libelle_francais', mark as unsolvable\n"," # print(f\"'libelle_francais' value not found : for index {index} : no matching 'libelle_francais' value\")\n"," index_not_found.append(index)\n","\n","print(f\"Amount of 'libelle_francais' found: {len(index_found)}\")\n","print(f\"Amount of 'libelle_francais' not found: {len(index_not_found)}\")"]},{"cell_type":"code","execution_count":69,"id":"m3S60h5Z_Dip","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":423},"executionInfo":{"elapsed":13,"status":"ok","timestamp":1737062660069,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"m3S60h5Z_Dip","outputId":"01ddfa92-6482-4051-97ed-a30f427c7ef2"},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" genre espece variete libelle_francais\n","528 Crataegus japonica NaN NaN\n","5282 Sorbus aria Pissardii' NaN\n","8900 Pyrus Fruit n. sp. Chanticleer' NaN\n","12375 Corylus cerasifera NaN NaN\n","12959 Parrotia persica Chanticleer' NaN\n","... ... ... ... ...\n","199405 Crataegus speciosa NaN NaN\n","199984 Malus purpurea NaN NaN\n","199993 Malus sargentii NaN NaN\n","199996 Malus sargentii NaN NaN\n","200128 Parrotia n. sp. NaN NaN\n","\n","[162 rows x 4 columns]"],"text/html":["\n"," <div id=\"df-aaa22710-e149-4331-8d05-0cac280fd09c\" class=\"colab-df-container\">\n"," <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>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>libelle_francais</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>528</th>\n"," <td>Crataegus</td>\n"," <td>japonica</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>5282</th>\n"," <td>Sorbus</td>\n"," <td>aria</td>\n"," <td>Pissardii'</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>8900</th>\n"," <td>Pyrus</td>\n"," <td>Fruit n. sp.</td>\n"," <td>Chanticleer'</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>12375</th>\n"," <td>Corylus</td>\n"," <td>cerasifera</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>12959</th>\n"," <td>Parrotia</td>\n"," <td>persica</td>\n"," <td>Chanticleer'</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>...</th>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>199405</th>\n"," <td>Crataegus</td>\n"," <td>speciosa</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>199984</th>\n"," <td>Malus</td>\n"," <td>purpurea</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>199993</th>\n"," <td>Malus</td>\n"," <td>sargentii</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>199996</th>\n"," <td>Malus</td>\n"," <td>sargentii</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," <tr>\n"," <th>200128</th>\n"," <td>Parrotia</td>\n"," <td>n. sp.</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>162 rows × 4 columns</p>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-aaa22710-e149-4331-8d05-0cac280fd09c')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-aaa22710-e149-4331-8d05-0cac280fd09c button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-aaa22710-e149-4331-8d05-0cac280fd09c');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-13660d0e-4b8b-4a2e-91a4-e0a3dc70813a\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-13660d0e-4b8b-4a2e-91a4-e0a3dc70813a')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-13660d0e-4b8b-4a2e-91a4-e0a3dc70813a button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_ffa495bc-5e9e-4468-8d9f-99818ed9fc40\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('remaining_cases')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_ffa495bc-5e9e-4468-8d9f-99818ed9fc40 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('remaining_cases');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"remaining_cases","repr_error":"0"}},"metadata":{},"execution_count":69}],"source":["remaining_cases = df_refined.loc[index_not_found][['genre','espece','variete','libelle_francais']].copy()\n","remaining_cases"]},{"cell_type":"code","execution_count":70,"id":"En8bvFMcBWt3","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":12,"status":"ok","timestamp":1737062660069,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"En8bvFMcBWt3","outputId":"a4ca5c6e-d700-41cf-8139-3f3550ad1c0c"},"outputs":[{"output_type":"stream","name":"stdout","text":["Different remaining cases : 51\n"," genre espece variete libelle_francais\n","12375 Corylus cerasifera NaN NaN\n","123822 Corylus pseudoacacia Unifolia' NaN\n","528 Crataegus japonica NaN NaN\n","199405 Crataegus speciosa NaN NaN\n","194059 Crataegus NaN NaN NaN\n","160955 Fraxinus betulus Fastigiata' NaN\n","161675 Fraxinus betulus NaN NaN\n","107494 Fraxinus calleryana Chanticleer' NaN\n","199101 Fraxinus chinensis NaN NaN\n","176284 Fraxinus minor NaN NaN\n","199154 Fraxinus pensylvanicum NaN NaN\n","198339 Fraxinus platyphyllos NaN NaN\n","108830 Fraxinus pseudoacacia Unifolia' NaN\n","94873 Fraxinus tulipifera NaN NaN\n","191688 Fraxinus NaN NaN NaN\n","141292 Gymnocladus NaN NaN NaN\n","170317 Ligustrum laevigata NaN NaN\n","194061 Ligustrum NaN NaN NaN\n","176508 Malus minor NaN NaN\n","80830 Malus nigra NaN NaN\n","197049 Malus padus NaN NaN\n","194678 Malus pumila Shinseiki' NaN\n","199984 Malus purpurea NaN NaN\n","199993 Malus sargentii NaN NaN\n","15330 Malus siliquastrum NaN NaN\n","19383 Malus NaN NaN NaN\n","15882 Ostrya avium NaN NaN\n","114634 Parrotia n. sp. NaN NaN\n","129676 Parrotia persica Baumannii' NaN\n","12959 Parrotia persica Chanticleer' NaN\n","80093 Parrotia sylvatica NaN NaN\n","73591 Prunus bignonioides NaN NaN\n","107662 Prunus calleryana Chanticleer' NaN\n","192112 Prunus cordata NaN NaN\n","191849 Prunus negundo NaN NaN\n","156348 Prunus pseudoacacia Unifolia' NaN\n","187246 Prunus pseudoplatanus NaN NaN\n","69885 Prunus NaN NaN NaN\n","8900 Pyrus Fruit n. sp. Chanticleer' NaN\n","146565 Pyrus alba Raket' NaN\n","166560 Pyrus australis NaN NaN\n","95678 Pyrus carpinifolia NaN NaN\n","107892 Pyrus controversa NaN NaN\n","187928 Pyrus domestica Bon chrétien Williams' NaN\n","193962 Pyrus domestica Drap d'Or' NaN\n","82969 Pyrus negundo NaN NaN\n","108675 Pyrus NaN NaN NaN\n","5282 Sorbus aria Pissardii' NaN\n","197045 Sorbus monspessulanum NaN NaN\n","15366 Sorbus tomentosa NaN NaN\n","195404 NaN NaN NaN NaN\n"]}],"source":["distinct_remaining_cases = remaining_cases.drop_duplicates().sort_values(by=['genre','espece','variete'])\n","print(f'Different remaining cases : {len(distinct_remaining_cases)}')\n","print(distinct_remaining_cases)\n","\n","# could go further but we will stop here"]},{"cell_type":"code","execution_count":71,"id":"OCR53Q0D1SaT","metadata":{"executionInfo":{"elapsed":7,"status":"ok","timestamp":1737062660069,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"OCR53Q0D1SaT"},"outputs":[],"source":["# some analysis and nan replacement need to be done on espece and variete\n","# impossible to have variete if only have genre and espece... Can take very long time"]},{"cell_type":"markdown","source":["#### espece"],"metadata":{"id":"jFJCBiY03t72"},"id":"jFJCBiY03t72"},{"cell_type":"code","source":["# prompt: take all the rows where \"espece\" is equal to nan. Then select the cases when variete is not nan\n","\n","# Assuming df_refined is your DataFrame\n","espece_nan_rows = df_refined[df_refined['espece'].isna()]\n","variete_not_nan_rows = espece_nan_rows[espece_nan_rows['variete'].notna()]\n","variete_not_nan_rows\n","\n","# same approach, based on finding the value based on the other fields. If possible, find cases where other both other fields and value are given"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":755},"id":"hB-pqw4S3epD","executionInfo":{"status":"ok","timestamp":1737062660069,"user_tz":-60,"elapsed":7,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"7c1b74b6-f5ca-41dd-c098-803579e43829"},"id":"hB-pqw4S3epD","execution_count":72,"outputs":[{"output_type":"execute_result","data":{"text/plain":[" id type_emplacement domanialite arrondissement \\\n","1725 101858 Arbre Jardin PARIS 19E ARRDT \n","2323 102582 Arbre Jardin PARIS 16E ARRDT \n","2733 103215 Arbre Jardin PARIS 17E ARRDT \n","2742 103224 Arbre Jardin PARIS 17E ARRDT \n","3944 104827 Arbre Jardin PARIS 13E ARRDT \n","... ... ... ... ... \n","198666 2022738 Arbre Alignement PARIS 11E ARRDT \n","198667 2022739 Arbre Alignement PARIS 11E ARRDT \n","198668 2022740 Arbre Alignement PARIS 11E ARRDT \n","198680 2022752 Arbre Alignement PARIS 11E ARRDT \n","198681 2022754 Arbre Alignement PARIS 11E ARRDT \n","\n"," complement_addresse numero \\\n","1725 NaN NaN \n","2323 82 à 90 NaN \n","2733 NaN NaN \n","2742 NaN NaN \n","3944 NaN NaN \n","... ... ... \n","198666 NaN NaN \n","198667 NaN NaN \n","198668 NaN NaN \n","198680 NaN NaN \n","198681 NaN NaN \n","\n"," lieu id_emplacement \\\n","1725 PARC DES BUTTES CHAUMONT D0030002 \n","2323 JARDIN DE L AVENUE FOCH / 10 AVENUE FOCH 408032 \n","2733 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000B0167 \n","2742 PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING 000B0188 \n","3944 SQUARE RENE LE GALL 10255 \n","... ... ... \n","198666 PLACE JEAN FERRAT 301001 \n","198667 PLACE JEAN FERRAT 301002 \n","198668 PLACE JEAN FERRAT 301003 \n","198680 PLACE JEAN FERRAT 105001 \n","198681 BOULEVARD DE BELLEVILLE 101001 \n","\n"," libelle_francais genre espece variete \\\n","1725 Pommier à fleurs Malus NaN Aldenhamensis' \n","2323 Cornouiller Cornus NaN Eddie's White Wonder' \n","2733 Cerisier à fleurs Prunus NaN Accolade' \n","2742 Cerisier à fleurs Prunus NaN Accolade' \n","3944 Pommier à fleurs Malus NaN Golden Hornet' \n","... ... ... ... ... \n","198666 Orme Ulmus NaN Nanguen' LUTECE \n","198667 Orme Ulmus NaN Nanguen' LUTECE \n","198668 Orme Ulmus NaN Nanguen' LUTECE \n","198680 Orme Ulmus NaN Nanguen' LUTECE \n","198681 Orme Ulmus NaN Nanguen' LUTECE \n","\n"," circonference_cm hauteur_m stade_developpement remarquable \\\n","1725 40.00 4.00 NaN NaN \n","2323 25.00 3.00 JA 0.00 \n","2733 32.00 NaN NaN NaN \n","2742 60.00 NaN NaN NaN \n","3944 40.00 4.00 NaN NaN \n","... ... ... ... ... \n","198666 30.00 5.00 NaN 0.00 \n","198667 20.00 5.00 NaN 0.00 \n","198668 20.00 5.00 NaN 0.00 \n","198680 20.00 5.00 NaN 0.00 \n","198681 35.00 7.00 NaN 0.00 \n","\n"," geo_point_2d_a geo_point_2d_b \n","1725 48.88 2.39 \n","2323 48.87 2.28 \n","2733 48.89 2.32 \n","2742 48.89 2.32 \n","3944 48.83 2.35 \n","... ... ... \n","198666 48.87 2.38 \n","198667 48.87 2.38 \n","198668 48.87 2.38 \n","198680 48.87 2.38 \n","198681 48.87 2.38 \n","\n","[1389 rows x 18 columns]"],"text/html":["\n"," <div id=\"df-99f43294-1d78-4259-aeb0-5cd124092b80\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>type_emplacement</th>\n"," <th>domanialite</th>\n"," <th>arrondissement</th>\n"," <th>complement_addresse</th>\n"," <th>numero</th>\n"," <th>lieu</th>\n"," <th>id_emplacement</th>\n"," <th>libelle_francais</th>\n"," <th>genre</th>\n"," <th>espece</th>\n"," <th>variete</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>stade_developpement</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>1725</th>\n"," <td>101858</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 19E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC DES BUTTES CHAUMONT</td>\n"," <td>D0030002</td>\n"," <td>Pommier à fleurs</td>\n"," <td>Malus</td>\n"," <td>NaN</td>\n"," <td>Aldenhamensis'</td>\n"," <td>40.00</td>\n"," <td>4.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.88</td>\n"," <td>2.39</td>\n"," </tr>\n"," <tr>\n"," <th>2323</th>\n"," <td>102582</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 16E ARRDT</td>\n"," <td>82 à 90</td>\n"," <td>NaN</td>\n"," <td>JARDIN DE L AVENUE FOCH / 10 AVENUE FOCH</td>\n"," <td>408032</td>\n"," <td>Cornouiller</td>\n"," <td>Cornus</td>\n"," <td>NaN</td>\n"," <td>Eddie's White Wonder'</td>\n"," <td>25.00</td>\n"," <td>3.00</td>\n"," <td>JA</td>\n"," <td>0.00</td>\n"," <td>48.87</td>\n"," <td>2.28</td>\n"," </tr>\n"," <tr>\n"," <th>2733</th>\n"," <td>103215</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000B0167</td>\n"," <td>Cerisier à fleurs</td>\n"," <td>Prunus</td>\n"," <td>NaN</td>\n"," <td>Accolade'</td>\n"," <td>32.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>2742</th>\n"," <td>103224</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 17E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PARC CLICHY-BATIGNOLLES-MARTIN LUTHER KING</td>\n"," <td>000B0188</td>\n"," <td>Cerisier à fleurs</td>\n"," <td>Prunus</td>\n"," <td>NaN</td>\n"," <td>Accolade'</td>\n"," <td>60.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.89</td>\n"," <td>2.32</td>\n"," </tr>\n"," <tr>\n"," <th>3944</th>\n"," <td>104827</td>\n"," <td>Arbre</td>\n"," <td>Jardin</td>\n"," <td>PARIS 13E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>SQUARE RENE LE GALL</td>\n"," <td>10255</td>\n"," <td>Pommier à fleurs</td>\n"," <td>Malus</td>\n"," <td>NaN</td>\n"," <td>Golden Hornet'</td>\n"," <td>40.00</td>\n"," <td>4.00</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>48.83</td>\n"," <td>2.35</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"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," <td>...</td>\n"," </tr>\n"," <tr>\n"," <th>198666</th>\n"," <td>2022738</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 11E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PLACE JEAN FERRAT</td>\n"," <td>301001</td>\n"," <td>Orme</td>\n"," <td>Ulmus</td>\n"," <td>NaN</td>\n"," <td>Nanguen' LUTECE</td>\n"," <td>30.00</td>\n"," <td>5.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.87</td>\n"," <td>2.38</td>\n"," </tr>\n"," <tr>\n"," <th>198667</th>\n"," <td>2022739</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 11E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PLACE JEAN FERRAT</td>\n"," <td>301002</td>\n"," <td>Orme</td>\n"," <td>Ulmus</td>\n"," <td>NaN</td>\n"," <td>Nanguen' LUTECE</td>\n"," <td>20.00</td>\n"," <td>5.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.87</td>\n"," <td>2.38</td>\n"," </tr>\n"," <tr>\n"," <th>198668</th>\n"," <td>2022740</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 11E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PLACE JEAN FERRAT</td>\n"," <td>301003</td>\n"," <td>Orme</td>\n"," <td>Ulmus</td>\n"," <td>NaN</td>\n"," <td>Nanguen' LUTECE</td>\n"," <td>20.00</td>\n"," <td>5.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.87</td>\n"," <td>2.38</td>\n"," </tr>\n"," <tr>\n"," <th>198680</th>\n"," <td>2022752</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 11E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>PLACE JEAN FERRAT</td>\n"," <td>105001</td>\n"," <td>Orme</td>\n"," <td>Ulmus</td>\n"," <td>NaN</td>\n"," <td>Nanguen' LUTECE</td>\n"," <td>20.00</td>\n"," <td>5.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.87</td>\n"," <td>2.38</td>\n"," </tr>\n"," <tr>\n"," <th>198681</th>\n"," <td>2022754</td>\n"," <td>Arbre</td>\n"," <td>Alignement</td>\n"," <td>PARIS 11E ARRDT</td>\n"," <td>NaN</td>\n"," <td>NaN</td>\n"," <td>BOULEVARD DE BELLEVILLE</td>\n"," <td>101001</td>\n"," <td>Orme</td>\n"," <td>Ulmus</td>\n"," <td>NaN</td>\n"," <td>Nanguen' LUTECE</td>\n"," <td>35.00</td>\n"," <td>7.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>48.87</td>\n"," <td>2.38</td>\n"," </tr>\n"," </tbody>\n","</table>\n","<p>1389 rows × 18 columns</p>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-99f43294-1d78-4259-aeb0-5cd124092b80')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-99f43294-1d78-4259-aeb0-5cd124092b80 button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-99f43294-1d78-4259-aeb0-5cd124092b80');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-79457e74-923f-4b6e-bbe5-da966eb7ef4b\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-79457e74-923f-4b6e-bbe5-da966eb7ef4b')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-79457e74-923f-4b6e-bbe5-da966eb7ef4b button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," <div id=\"id_e0749d19-0f61-467d-b46b-c935e7d0a971\">\n"," <style>\n"," .colab-df-generate {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-generate:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," [theme=dark] .colab-df-generate {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-generate:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n"," <button class=\"colab-df-generate\" onclick=\"generateWithVariable('variete_not_nan_rows')\"\n"," title=\"Generate code using this dataframe.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n"," </svg>\n"," </button>\n"," <script>\n"," (() => {\n"," const buttonEl =\n"," document.querySelector('#id_e0749d19-0f61-467d-b46b-c935e7d0a971 button.colab-df-generate');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," buttonEl.onclick = () => {\n"," google.colab.notebook.generateWithVariable('variete_not_nan_rows');\n"," }\n"," })();\n"," </script>\n"," </div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","variable_name":"variete_not_nan_rows","repr_error":"0"}},"metadata":{},"execution_count":72}]},{"cell_type":"markdown","metadata":{"id":"32tkfElqDL3G"},"source":["### Before and After cleaning anomalies"],"id":"32tkfElqDL3G"},{"cell_type":"code","source":["import seaborn as sns\n","\n","\n","# Calculate the number of non-missing values for each column\n","non_missing_original = df.notnull().sum()\n","non_missing_refined = df_refined.notnull().sum()\n","\n","# Prepare data for the bar plot\n","comparison_data = pd.DataFrame({\n"," 'Columns': non_missing_original.index,\n"," 'Original Dataset': non_missing_original.values,\n"," 'Corrected Dataset': non_missing_refined.values\n","}).melt(id_vars='Columns', var_name='Dataset', value_name='Non-Missing Count')\n","\n","\n","# Create the bar plot\n","plt.figure(figsize=(12, 6))\n","ax = sns.barplot(\n"," data=comparison_data,\n"," x='Columns',\n"," y='Non-Missing Count',\n"," hue='Dataset',\n"," palette='pastel'\n",")\n","# Rotate the x-axis labels\n","plt.xticks(rotation=45, ha='right')\n","\n","# Add count annotations to each bar (inside the bars)\n","for p in ax.patches:\n"," height = p.get_height() # Bar height (value of the bar)\n"," if height > 0: # Only annotate bars with positive height\n"," ax.annotate(\n"," f'{int(height)}', # Display the count as an integer\n"," (p.get_x() + p.get_width() / 2., height / 2), # Position inside the bar, vertically centered\n"," ha='center', va='center', fontsize=10, color='black', rotation=90\n"," )\n","\n","# Add plot titles and labels\n","plt.title('Comparison of Non-Missing Values by Column', fontsize=14)\n","plt.ylabel('Number of Non-Missing Values', fontsize=12)\n","plt.xlabel('Columns', fontsize=12)\n","plt.legend(title='Dataset', loc='upper right')\n","plt.grid(axis='y', linestyle='--', alpha=0.7)\n","plt.tight_layout() # Adjust layout to prevent overlap\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":573},"id":"1PMO745VZod2","executionInfo":{"status":"ok","timestamp":1737062662168,"user_tz":-60,"elapsed":1591,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"1b7e704a-78d0-43e6-9d0c-b47209c7d69e"},"execution_count":73,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1200x600 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"id":"1PMO745VZod2"},{"cell_type":"code","source":["# Create a DataFrame for plotting\n","diff_df = data_diff = (non_missing_refined - non_missing_original).reset_index()\n","diff_df.columns = ['Columns', 'Difference']\n","\n","# Plot the data using Seaborn\n","plt.figure(figsize=(10, 6))\n","ax = sns.barplot(\n"," data=diff_df,\n"," x=diff_df.columns[0],\n"," y=diff_df.columns[1],\n"," palette=[\"blue\" if x > 0 else \"red\" for x in diff_df['Difference']] # Blue for positive, Red for negative\n",")\n","\n","# Set the y-axis to be centered at 0\n","ax.axhline(0, color='black',linewidth=1)\n","\n","# Add plot titles and labels\n","plt.title('Differences Between Original and Corrected Datasets counts', fontsize=14)\n","plt.ylabel('Difference (Original - Corrected)', fontsize=12)\n","plt.xlabel('Columns', fontsize=12)\n","\n","# Rotate the x-axis labels\n","plt.xticks(rotation=45, ha='right')\n","\n","# Display the values inside the bars\n","for p in ax.patches:\n"," height = p.get_height()\n"," ax.text(\n"," p.get_x() + p.get_width() / 2, # Position at the center of the bar\n"," height/2, # Position at the top of the bar\n"," f'{int(height)}', # Annotate with the number\n"," ha='center', va='center', fontsize=10, color='k', rotation=0\n"," )\n","\n","# Show the plot\n","plt.tight_layout() # Adjust the layout to avoid overlap\n","plt.show()"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":717},"id":"5D75sJRqs4vW","executionInfo":{"status":"ok","timestamp":1737062662968,"user_tz":-60,"elapsed":815,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"8573649c-9423-4cd1-ea41-888d00186431"},"execution_count":74,"outputs":[{"output_type":"stream","name":"stderr","text":["<ipython-input-74-1a4a04441649>:7: FutureWarning: \n","\n","Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n","\n"," ax = sns.barplot(\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x600 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"id":"5D75sJRqs4vW"},{"cell_type":"code","execution_count":75,"metadata":{"executionInfo":{"elapsed":910,"status":"ok","timestamp":1737062663874,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"8Tx4l94HDGb2","colab":{"base_uri":"https://localhost:8080/","height":849},"outputId":"508580f5-c834-4ec2-9e37-ebc54d7e8d0d"},"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}],"source":["# boxplots\n","for variable in quantitative_variables:\n"," plot_univariate_variable_analysis(df_refined, variable, 'boxplot')"],"id":"8Tx4l94HDGb2"},{"cell_type":"markdown","id":"V7hc3wnUW7gq","metadata":{"id":"V7hc3wnUW7gq"},"source":["## 4) Analysis"]},{"cell_type":"markdown","id":"uZfu0OOUYRI3","metadata":{"id":"uZfu0OOUYRI3"},"source":["### Naive analysis (after data clean)"]},{"cell_type":"code","execution_count":76,"id":"2SUutH-hYK1E","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":7,"status":"ok","timestamp":1737062663874,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"2SUutH-hYK1E","outputId":"6c8e2042-bb7c-4da3-e17f-049f01c6632b","scrolled":true},"outputs":[{"output_type":"stream","name":"stdout","text":["<class 'pandas.core.frame.DataFrame'>\n","RangeIndex: 200137 entries, 0 to 200136\n","Data columns (total 18 columns):\n"," # Column Non-Null Count Dtype \n","--- ------ -------------- ----- \n"," 0 id 200137 non-null int64 \n"," 1 type_emplacement 200137 non-null object \n"," 2 domanialite 200137 non-null object \n"," 3 arrondissement 200137 non-null object \n"," 4 complement_addresse 30902 non-null object \n"," 5 numero 0 non-null float64\n"," 6 lieu 200137 non-null object \n"," 7 id_emplacement 200137 non-null object \n"," 8 libelle_francais 199975 non-null object \n"," 9 genre 200121 non-null object \n"," 10 espece 198385 non-null object \n"," 11 variete 36777 non-null object \n"," 12 circonference_cm 199514 non-null float64\n"," 13 hauteur_m 186181 non-null float64\n"," 14 stade_developpement 132932 non-null object \n"," 15 remarquable 137039 non-null float64\n"," 16 geo_point_2d_a 200137 non-null float64\n"," 17 geo_point_2d_b 200137 non-null float64\n","dtypes: float64(6), int64(1), object(11)\n","memory usage: 27.5+ MB\n"]}],"source":["df_refined.info() # display count of columns, lines and columns names"]},{"cell_type":"code","execution_count":77,"id":"X0QJb_HCYK1E","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":300},"executionInfo":{"elapsed":6,"status":"ok","timestamp":1737062663874,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"X0QJb_HCYK1E","outputId":"a67442cd-a0c3-4598-d61f-53d7baae71c9","scrolled":true},"outputs":[{"output_type":"execute_result","data":{"text/plain":[" id numero circonference_cm hauteur_m remarquable \\\n","count 200137.00 0.00 199514.00 186181.00 137039.00 \n","mean 387202.68 NaN 80.26 8.93 0.00 \n","std 545603.24 NaN 67.48 5.97 0.04 \n","min 99874.00 NaN 0.00 0.00 0.00 \n","25% 155927.00 NaN 30.00 5.00 0.00 \n","50% 221078.00 NaN 70.00 9.00 0.00 \n","75% 274102.00 NaN 115.00 12.00 0.00 \n","max 2024745.00 NaN 3120.00 51.55 1.00 \n","\n"," geo_point_2d_a geo_point_2d_b \n","count 200137.00 200137.00 \n","mean 48.85 2.35 \n","std 0.03 0.05 \n","min 48.74 2.21 \n","25% 48.84 2.31 \n","50% 48.85 2.35 \n","75% 48.88 2.39 \n","max 48.91 2.47 "],"text/html":["\n"," <div id=\"df-e04d59a7-eb12-48a1-bceb-e05a6188120f\" class=\"colab-df-container\">\n"," <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>id</th>\n"," <th>numero</th>\n"," <th>circonference_cm</th>\n"," <th>hauteur_m</th>\n"," <th>remarquable</th>\n"," <th>geo_point_2d_a</th>\n"," <th>geo_point_2d_b</th>\n"," </tr>\n"," </thead>\n"," <tbody>\n"," <tr>\n"," <th>count</th>\n"," <td>200137.00</td>\n"," <td>0.00</td>\n"," <td>199514.00</td>\n"," <td>186181.00</td>\n"," <td>137039.00</td>\n"," <td>200137.00</td>\n"," <td>200137.00</td>\n"," </tr>\n"," <tr>\n"," <th>mean</th>\n"," <td>387202.68</td>\n"," <td>NaN</td>\n"," <td>80.26</td>\n"," <td>8.93</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.35</td>\n"," </tr>\n"," <tr>\n"," <th>std</th>\n"," <td>545603.24</td>\n"," <td>NaN</td>\n"," <td>67.48</td>\n"," <td>5.97</td>\n"," <td>0.04</td>\n"," <td>0.03</td>\n"," <td>0.05</td>\n"," </tr>\n"," <tr>\n"," <th>min</th>\n"," <td>99874.00</td>\n"," <td>NaN</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>0.00</td>\n"," <td>48.74</td>\n"," <td>2.21</td>\n"," </tr>\n"," <tr>\n"," <th>25%</th>\n"," <td>155927.00</td>\n"," <td>NaN</td>\n"," <td>30.00</td>\n"," <td>5.00</td>\n"," <td>0.00</td>\n"," <td>48.84</td>\n"," <td>2.31</td>\n"," </tr>\n"," <tr>\n"," <th>50%</th>\n"," <td>221078.00</td>\n"," <td>NaN</td>\n"," <td>70.00</td>\n"," <td>9.00</td>\n"," <td>0.00</td>\n"," <td>48.85</td>\n"," <td>2.35</td>\n"," </tr>\n"," <tr>\n"," <th>75%</th>\n"," <td>274102.00</td>\n"," <td>NaN</td>\n"," <td>115.00</td>\n"," <td>12.00</td>\n"," <td>0.00</td>\n"," <td>48.88</td>\n"," <td>2.39</td>\n"," </tr>\n"," <tr>\n"," <th>max</th>\n"," <td>2024745.00</td>\n"," <td>NaN</td>\n"," <td>3120.00</td>\n"," <td>51.55</td>\n"," <td>1.00</td>\n"," <td>48.91</td>\n"," <td>2.47</td>\n"," </tr>\n"," </tbody>\n","</table>\n","</div>\n"," <div class=\"colab-df-buttons\">\n","\n"," <div class=\"colab-df-container\">\n"," <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e04d59a7-eb12-48a1-bceb-e05a6188120f')\"\n"," title=\"Convert this dataframe to an interactive table.\"\n"," style=\"display:none;\">\n","\n"," <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n"," <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n"," </svg>\n"," </button>\n","\n"," <style>\n"," .colab-df-container {\n"," display:flex;\n"," gap: 12px;\n"," }\n","\n"," .colab-df-convert {\n"," background-color: #E8F0FE;\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: #1967D2;\n"," height: 32px;\n"," padding: 0 0 0 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-convert:hover {\n"," background-color: #E2EBFA;\n"," box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: #174EA6;\n"," }\n","\n"," .colab-df-buttons div {\n"," margin-bottom: 4px;\n"," }\n","\n"," [theme=dark] .colab-df-convert {\n"," background-color: #3B4455;\n"," fill: #D2E3FC;\n"," }\n","\n"," [theme=dark] .colab-df-convert:hover {\n"," background-color: #434B5C;\n"," box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n"," filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n"," fill: #FFFFFF;\n"," }\n"," </style>\n","\n"," <script>\n"," const buttonEl =\n"," document.querySelector('#df-e04d59a7-eb12-48a1-bceb-e05a6188120f button.colab-df-convert');\n"," buttonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n","\n"," async function convertToInteractive(key) {\n"," const element = document.querySelector('#df-e04d59a7-eb12-48a1-bceb-e05a6188120f');\n"," const dataTable =\n"," await google.colab.kernel.invokeFunction('convertToInteractive',\n"," [key], {});\n"," if (!dataTable) return;\n","\n"," const docLinkHtml = 'Like what you see? Visit the ' +\n"," '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n"," + ' to learn more about interactive tables.';\n"," element.innerHTML = '';\n"," dataTable['output_type'] = 'display_data';\n"," await google.colab.output.renderOutput(dataTable, element);\n"," const docLink = document.createElement('div');\n"," docLink.innerHTML = docLinkHtml;\n"," element.appendChild(docLink);\n"," }\n"," </script>\n"," </div>\n","\n","\n","<div id=\"df-0a9428b7-449c-42e6-bb88-a36512b95f6c\">\n"," <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-0a9428b7-449c-42e6-bb88-a36512b95f6c')\"\n"," title=\"Suggest charts\"\n"," style=\"display:none;\">\n","\n","<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n"," width=\"24px\">\n"," <g>\n"," <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n"," </g>\n","</svg>\n"," </button>\n","\n","<style>\n"," .colab-df-quickchart {\n"," --bg-color: #E8F0FE;\n"," --fill-color: #1967D2;\n"," --hover-bg-color: #E2EBFA;\n"," --hover-fill-color: #174EA6;\n"," --disabled-fill-color: #AAA;\n"," --disabled-bg-color: #DDD;\n"," }\n","\n"," [theme=dark] .colab-df-quickchart {\n"," --bg-color: #3B4455;\n"," --fill-color: #D2E3FC;\n"," --hover-bg-color: #434B5C;\n"," --hover-fill-color: #FFFFFF;\n"," --disabled-bg-color: #3B4455;\n"," --disabled-fill-color: #666;\n"," }\n","\n"," .colab-df-quickchart {\n"," background-color: var(--bg-color);\n"," border: none;\n"," border-radius: 50%;\n"," cursor: pointer;\n"," display: none;\n"," fill: var(--fill-color);\n"," height: 32px;\n"," padding: 0;\n"," width: 32px;\n"," }\n","\n"," .colab-df-quickchart:hover {\n"," background-color: var(--hover-bg-color);\n"," box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n"," fill: var(--button-hover-fill-color);\n"," }\n","\n"," .colab-df-quickchart-complete:disabled,\n"," .colab-df-quickchart-complete:disabled:hover {\n"," background-color: var(--disabled-bg-color);\n"," fill: var(--disabled-fill-color);\n"," box-shadow: none;\n"," }\n","\n"," .colab-df-spinner {\n"," border: 2px solid var(--fill-color);\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," animation:\n"," spin 1s steps(1) infinite;\n"," }\n","\n"," @keyframes spin {\n"," 0% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," border-left-color: var(--fill-color);\n"," }\n"," 20% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 30% {\n"," border-color: transparent;\n"," border-left-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," border-right-color: var(--fill-color);\n"," }\n"," 40% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-top-color: var(--fill-color);\n"," }\n"," 60% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," }\n"," 80% {\n"," border-color: transparent;\n"," border-right-color: var(--fill-color);\n"," border-bottom-color: var(--fill-color);\n"," }\n"," 90% {\n"," border-color: transparent;\n"," border-bottom-color: var(--fill-color);\n"," }\n"," }\n","</style>\n","\n"," <script>\n"," async function quickchart(key) {\n"," const quickchartButtonEl =\n"," document.querySelector('#' + key + ' button');\n"," quickchartButtonEl.disabled = true; // To prevent multiple clicks.\n"," quickchartButtonEl.classList.add('colab-df-spinner');\n"," try {\n"," const charts = await google.colab.kernel.invokeFunction(\n"," 'suggestCharts', [key], {});\n"," } catch (error) {\n"," console.error('Error during call to suggestCharts:', error);\n"," }\n"," quickchartButtonEl.classList.remove('colab-df-spinner');\n"," quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n"," }\n"," (() => {\n"," let quickchartButtonEl =\n"," document.querySelector('#df-0a9428b7-449c-42e6-bb88-a36512b95f6c button');\n"," quickchartButtonEl.style.display =\n"," google.colab.kernel.accessAllowed ? 'block' : 'none';\n"," })();\n"," </script>\n","</div>\n","\n"," </div>\n"," </div>\n"],"application/vnd.google.colaboratory.intrinsic+json":{"type":"dataframe","summary":"{\n \"name\": \"## > 250000 cm as circumference, > 800000 m as height\",\n \"rows\": 8,\n \"fields\": [\n {\n \"column\": \"id\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 636490.3531323484,\n \"min\": 99874.0,\n \"max\": 2024745.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 387202.68230761925,\n 221078.0,\n 200137.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"numero\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": 0.0,\n \"max\": 0.0,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"circonference_cm\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 70371.0990261856,\n \"min\": 0.0,\n \"max\": 199514.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 80.26401656024139\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"hauteur_m\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 65820.25619681984,\n \"min\": 0.0,\n \"max\": 186181.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 8.933927092453045\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"remarquable\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 48450.55066982106,\n \"min\": 0.0,\n \"max\": 137039.0,\n \"num_unique_values\": 5,\n \"samples\": [\n 0.0013426834696692183\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_a\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 70744.31300132743,\n \"min\": 0.0302341454886933,\n \"max\": 200137.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 48.85449078901268\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geo_point_2d_b\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 70758.40152298885,\n \"min\": 0.05121963963079915,\n \"max\": 200137.0,\n \"num_unique_values\": 8,\n \"samples\": [\n 2.3482078955462966\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}"}},"metadata":{},"execution_count":77}],"source":["df_refined.describe()# general statistics only on number types variables (int and floats)\n","\n","# problems:\n","## 0 as min value for circumference, height\n","## > 250000 cm as circumference, > 800000 m as height"]},{"cell_type":"code","execution_count":78,"id":"02rOaFuYYrxA","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":418,"status":"ok","timestamp":1737062664287,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"02rOaFuYYrxA","outputId":"b3c3f06e-6f36-4a63-b85e-de27555009b0"},"outputs":[{"output_type":"stream","name":"stdout","text":["Number of different species: 539\n"]}],"source":["print(f\"Number of different species: {df_refined['espece'].nunique()}\")"]},{"cell_type":"code","execution_count":79,"id":"5vyyd7-lZ_tT","metadata":{"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":12,"status":"ok","timestamp":1737062664287,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"5vyyd7-lZ_tT","outputId":"56e93bbe-fc4d-4de0-ea00-8081b3dde942"},"outputs":[{"output_type":"stream","name":"stdout","text":["Number of remarquable different species: 66\n"]}],"source":["print(f\"Number of remarquable different species: {df_refined[df_refined['remarquable']==1]['espece'].nunique()}\")"]},{"cell_type":"markdown","id":"-859g19jW-0l","metadata":{"id":"-859g19jW-0l"},"source":["### Univariate Analysis"]},{"cell_type":"code","execution_count":80,"id":"_we2vsHTWuTn","metadata":{"executionInfo":{"elapsed":9,"status":"ok","timestamp":1737062664287,"user":{"displayName":"Grégoire","userId":"01249298206117462766"},"user_tz":-60},"id":"_we2vsHTWuTn"},"outputs":[],"source":["interesting_variables = {'domanialite':'pie','arrondissement':'heat_map',\n"," 'lieu':'heat_map', 'libelle_francais':'bars',\n"," 'genre':'bars', 'espece':'bars',\n"," 'variete':'bars','circonference_cm':'hist',\n"," 'hauteur_m':'hist','stade_developpement':'pie',\n"," 'remarquable':'bars_log'}"]},{"cell_type":"code","execution_count":81,"id":"tn6p_aQcXmLM","metadata":{"colab":{"base_uri":"https://localhost:8080/","height":1000,"output_embedded_package_id":"13LADlWprO2_fGI3yOouYyyu2hHs99ijc"},"id":"tn6p_aQcXmLM","outputId":"6152b149-0cdb-4d21-8390-8fb3fbaba709","executionInfo":{"status":"ok","timestamp":1737062696721,"user_tz":-60,"elapsed":32443,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[{"output_type":"display_data","data":{"text/plain":"Output hidden; open in https://colab.research.google.com to view."},"metadata":{}}],"source":["for variable in interesting_variables:\n"," plot_univariate_variable_analysis(df_refined, variable, interesting_variables[variable])\n"," print()"]},{"cell_type":"code","execution_count":82,"id":"58c0f807-803f-4cef-9ff6-a827020cb19f","metadata":{"id":"58c0f807-803f-4cef-9ff6-a827020cb19f","executionInfo":{"status":"ok","timestamp":1737062696721,"user_tz":-60,"elapsed":7,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}}},"outputs":[],"source":["# remains to do\n","## multivariate analysis (not yet)"]},{"cell_type":"code","source":["plot_bivariate_variable_analysis(df_refined, 'remarquable', 'hauteur_m', 'boxplot')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":564},"id":"OQrQVkhob6oR","executionInfo":{"status":"ok","timestamp":1737062696721,"user_tz":-60,"elapsed":7,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"fe96a611-1254-4c7d-fdc5-1ca04d6709e4"},"id":"OQrQVkhob6oR","execution_count":83,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 800x600 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["plot_bivariate_variable_analysis(df_refined, 'remarquable', 'circonference_cm', 'boxplot')"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":564},"id":"olxCasOYdXxO","executionInfo":{"status":"ok","timestamp":1737062697779,"user_tz":-60,"elapsed":1063,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"af497138-e681-46bb-aaff-b5065ffd4b76"},"id":"olxCasOYdXxO","execution_count":84,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 800x600 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["variable1_name = 'remarquable'\n","variable2_name = 'hauteur_m'\n","pearsonr, p_value = compute_bivariate_stats_indicators(df_refined, variable1_name, variable2_name)\n","print(f\"Pearson correlation coefficient between {variable1_name} and {variable2_name} : {pearsonr}\") # existing correlation but very weak\n","\n","print()\n","\n","variable1_name = 'remarquable'\n","variable2_name = 'circonference_cm'\n","pearsonr, p_value = compute_bivariate_stats_indicators(df_refined, variable1_name, variable2_name)\n","print(f\"Pearson correlation coefficient between {variable1_name} and {variable2_name} : {pearsonr}\") # existing correlation but very weak"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GtSIhb2Hdcsm","executionInfo":{"status":"ok","timestamp":1737062782320,"user_tz":-60,"elapsed":1522,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"8d1a664b-6b5f-4175-e597-70f236770a10"},"id":"GtSIhb2Hdcsm","execution_count":87,"outputs":[{"output_type":"stream","name":"stdout","text":["Both quantitative variables : Pearson correlation analysis\n","Pearson correlation coefficient between remarquable and hauteur_m : 0.06269852992286187\n","\n","Both quantitative variables : Pearson correlation analysis\n","Pearson correlation coefficient between remarquable and circonference_cm : 0.11668322959396166\n"]}]},{"cell_type":"code","source":["eta_squared, anova_table = compute_bivariate_stats_indicators(df_refined, 'remarquable','espece')\n","print(f\"Eta-squared: {eta_squared}\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"Rwt35nIpBJa2","executionInfo":{"status":"ok","timestamp":1737062823550,"user_tz":-60,"elapsed":37593,"user":{"displayName":"Grégoire","userId":"01249298206117462766"}},"outputId":"0aaeabc1-1de3-4c14-e7f9-a7ebb154cd81"},"id":"Rwt35nIpBJa2","execution_count":88,"outputs":[{"output_type":"stream","name":"stdout","text":["Both quantitative and categorical variables : ANOVA analysis\n","Eta-squared: 0.05512247910689265\n"]},{"output_type":"stream","name":"stderr","text":["/usr/local/lib/python3.10/dist-packages/statsmodels/base/model.py:1894: ValueWarning: covariance of constraints does not have full rank. The number of constraints is 538, but rank is 481\n"," warnings.warn('covariance of constraints does not have full '\n","/content/drive/MyDrive/Colab_Notebooks/Project_2/data_analysis_functions.py:295: FutureWarning: Series.__getitem__ treating keys as positions is deprecated. In a future version, integer keys will always be treated as labels (consistent with DataFrame behavior). To access a value by position, use `ser.iloc[pos]`\n"," eta_squared = anova_table['sum_sq'][0] / sum(anova_table['sum_sq'])\n"]}]},{"cell_type":"code","source":[],"metadata":{"id":"oGVLCs_pAqFi"},"id":"oGVLCs_pAqFi","execution_count":null,"outputs":[]}],"metadata":{"colab":{"provenance":[],"toc_visible":true},"kernelspec":{"display_name":"Python 3 (ipykernel)","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":"3.12.7"}},"nbformat":4,"nbformat_minor":5} |