.. _seance6graphesenoncerst:

================
Graphes - énoncé
================


.. only:: html

    **Links:** :download:`notebook <seance6_graphes_enonce.ipynb>`, :downloadlink:`html <seance6_graphes_enonce2html.html>`, :download:`PDF <seance6_graphes_enonce.pdf>`, :download:`python <seance6_graphes_enonce.py>`, :downloadlink:`slides <seance6_graphes_enonce.slides.html>`, :githublink:`GitHub|_doc/notebooks/sessions/seance6_graphes_enonce.ipynb|*`


Ce notebook introduit `matplotlib <http://matplotlib.org/>`__ et
d’autres modules Python qui permettent de tracer des graphes et bâtis
sur la même logique que matplotlib.

Pour avoir des graphiques inclus dans le notebook, il faut ajouter cette
ligne et l’exécuter en premier.

.. code:: ipython3

    %matplotlib inline

On change le style pour un style plus moderne, celui de
`ggplot <http://ggplot2.org/>`__ :

.. code:: ipython3

    import matplotlib.pyplot as plt
    plt.style.use('ggplot')

.. code:: ipython3

    from jyquickhelper import add_notebook_menu
    add_notebook_menu()






.. contents::
    :local:





Données
-------

Pour tous les exemples qui suivent, on utilise les résultat `élection
présidentielle de
2012 <https://www.data.gouv.fr/fr/datasets/election-presidentielle-2012-resultats-572124/>`__.
Si vous n’avez pas le module
`actuariat_python <http://www.xavierdupre.fr/app/actuariat_python/helpsphinx/index.html>`__,
il vous suffit de recopier le code de la fonction
`elections_presidentielles <http://www.xavierdupre.fr/app/actuariat_python/helpsphinx/_modules/actuariat_python/data/elections.html#elections_presidentielles>`__
qui utilise la fonction
`read_excel <http://pandas.pydata.org/pandas-docs/stable/generated/pandas.read_excel.html>`__.
La fonction utilise des données stockées localement afin que le code
ci-dessous fonctionne toujours même si le format des données change sur
le site *data.gouv.fr*.

.. code:: ipython3

    from actuariat_python.data import elections_presidentielles
    dict_df = elections_presidentielles(local=True, agg="dep")

.. code:: ipython3

    list(dict_df.keys())




.. parsed-literal::
    ['circ1', 'circ2', 'dep1', 'dep2']



.. code:: ipython3

    dict_df["dep1"].head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Code du département</th>
          <th>Libellé du département</th>
          <th>Code de la circonscription</th>
          <th>Inscrits</th>
          <th>Votants</th>
          <th>Exprimés</th>
          <th>Blancs et nuls</th>
          <th>Nathalie ARTHAUD (LO)</th>
          <th>Philippe POUTOU (NPA)</th>
          <th>Jean-Luc MELENCHON (FG)</th>
          <th>François HOLLANDE (PS)</th>
          <th>Eva JOLY (EELV)</th>
          <th>François BAYROU (MODEM)</th>
          <th>Nicolas SARKOZY (UMP)</th>
          <th>Nicolas DUPONT-AIGNAN (DLR)</th>
          <th>Marine LE PEN (FN)</th>
          <th>Jacques CHEMINADE (SP)</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>1</td>
          <td>AIN</td>
          <td>15</td>
          <td>393808</td>
          <td>327812</td>
          <td>321359</td>
          <td>6453</td>
          <td>1794</td>
          <td>3323</td>
          <td>30898</td>
          <td>73096</td>
          <td>7268</td>
          <td>32650</td>
          <td>97722</td>
          <td>7208</td>
          <td>66540</td>
          <td>860</td>
        </tr>
        <tr>
          <th>1</th>
          <td>2</td>
          <td>AISNE</td>
          <td>15</td>
          <td>376068</td>
          <td>303140</td>
          <td>297944</td>
          <td>5196</td>
          <td>2490</td>
          <td>3860</td>
          <td>30360</td>
          <td>80751</td>
          <td>3455</td>
          <td>19895</td>
          <td>72090</td>
          <td>5853</td>
          <td>78452</td>
          <td>738</td>
        </tr>
        <tr>
          <th>2</th>
          <td>3</td>
          <td>ALLIER</td>
          <td>6</td>
          <td>256275</td>
          <td>211009</td>
          <td>205950</td>
          <td>5059</td>
          <td>1482</td>
          <td>2584</td>
          <td>27969</td>
          <td>61131</td>
          <td>3232</td>
          <td>17814</td>
          <td>49477</td>
          <td>4068</td>
          <td>37736</td>
          <td>457</td>
        </tr>
        <tr>
          <th>3</th>
          <td>4</td>
          <td>ALPES-DE-HAUTE-PROVENCE</td>
          <td>3</td>
          <td>123933</td>
          <td>102899</td>
          <td>100788</td>
          <td>2111</td>
          <td>487</td>
          <td>1394</td>
          <td>15269</td>
          <td>24551</td>
          <td>2933</td>
          <td>7483</td>
          <td>25668</td>
          <td>1845</td>
          <td>20875</td>
          <td>283</td>
        </tr>
        <tr>
          <th>4</th>
          <td>5</td>
          <td>HAUTES-ALPES</td>
          <td>3</td>
          <td>106865</td>
          <td>88619</td>
          <td>86777</td>
          <td>1842</td>
          <td>488</td>
          <td>1152</td>
          <td>12175</td>
          <td>21248</td>
          <td>3147</td>
          <td>8559</td>
          <td>22655</td>
          <td>1782</td>
          <td>15359</td>
          <td>212</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    dict_df["dep2"].head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Code du département</th>
          <th>Libellé du département</th>
          <th>Code de la circonscription</th>
          <th>Inscrits</th>
          <th>Votants</th>
          <th>Exprimés</th>
          <th>Blancs et nuls</th>
          <th>François HOLLANDE (PS)</th>
          <th>Nicolas SARKOZY (UMP)</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>979</td>
          <td>SAINT-BARTHELEMY et SAINT-MARTIN</td>
          <td>1</td>
          <td>22686</td>
          <td>9907</td>
          <td>9492</td>
          <td>415</td>
          <td>3851</td>
          <td>5641</td>
        </tr>
        <tr>
          <th>1</th>
          <td>01</td>
          <td>AIN</td>
          <td>15</td>
          <td>393866</td>
          <td>326587</td>
          <td>307074</td>
          <td>19513</td>
          <td>131333</td>
          <td>175741</td>
        </tr>
        <tr>
          <th>2</th>
          <td>02</td>
          <td>AISNE</td>
          <td>15</td>
          <td>376073</td>
          <td>302076</td>
          <td>281020</td>
          <td>21056</td>
          <td>147260</td>
          <td>133760</td>
        </tr>
        <tr>
          <th>3</th>
          <td>03</td>
          <td>ALLIER</td>
          <td>6</td>
          <td>256211</td>
          <td>211132</td>
          <td>196208</td>
          <td>14924</td>
          <td>111615</td>
          <td>84593</td>
        </tr>
        <tr>
          <th>4</th>
          <td>04</td>
          <td>ALPES-DE-HAUTE-PROVENCE</td>
          <td>3</td>
          <td>123895</td>
          <td>103581</td>
          <td>96942</td>
          <td>6639</td>
          <td>49498</td>
          <td>47444</td>
        </tr>
      </tbody>
    </table>
    </div>



On corrige le code du département ``01`` –> ``1``.

.. code:: ipython3

    def cleandep(s):
        if isinstance(s, str):
             r = s.lstrip('0')
        else:
            r = str(s)
        return r
    dict_df["dep1"]["Code du département"] = dict_df["dep1"]["Code du département"].apply(cleandep)
    dict_df["dep2"]["Code du département"] = dict_df["dep2"]["Code du département"].apply(cleandep)

.. code:: ipython3

    deps = dict_df["dep1"].merge(dict_df["dep2"], 
                                           on="Code du département",
                                           suffixes=("T1", "T2"))
    deps.columns




.. parsed-literal::
    Index(['Code du département', 'Libellé du départementT1',
           'Code de la circonscriptionT1', 'InscritsT1', 'VotantsT1', 'ExprimésT1',
           'Blancs et nulsT1', 'Nathalie ARTHAUD (LO)', 'Philippe POUTOU (NPA)',
           'Jean-Luc MELENCHON (FG)', 'François HOLLANDE (PS)T1',
           'Eva JOLY (EELV)', 'François BAYROU (MODEM)', 'Nicolas SARKOZY (UMP)T1',
           'Nicolas DUPONT-AIGNAN (DLR)', 'Marine LE PEN (FN)',
           'Jacques CHEMINADE (SP)', 'Libellé du départementT2',
           'Code de la circonscriptionT2', 'InscritsT2', 'VotantsT2', 'ExprimésT2',
           'Blancs et nulsT2', 'François HOLLANDE (PS)T2',
           'Nicolas SARKOZY (UMP)T2'],
          dtype='object')



.. code:: ipython3

    deps["rHollandeT1"] = deps['François HOLLANDE (PS)T1'] / (deps["VotantsT1"] - deps["Blancs et nulsT1"])
    deps["rSarkozyT1"] = deps['Nicolas SARKOZY (UMP)T1'] / (deps["VotantsT1"] - deps["Blancs et nulsT1"])
    deps["rNulT1"] = deps["Blancs et nulsT1"] / deps["VotantsT1"]
    deps["rHollandeT2"] = deps["François HOLLANDE (PS)T2"] / (deps["VotantsT2"] - deps["Blancs et nulsT2"])
    deps["rSarkozyT2"] = deps['Nicolas SARKOZY (UMP)T2'] / (deps["VotantsT2"] - deps["Blancs et nulsT2"])
    deps["rNulT2"] = deps["Blancs et nulsT2"] / deps["VotantsT2"]
    data = deps[["Code du département", "Libellé du départementT1", 
                 "VotantsT1", "rHollandeT1", "rSarkozyT1", "rNulT1",
                 "VotantsT2", "rHollandeT2", "rSarkozyT2", "rNulT2"]]
    data_elections = data # parfois data est remplacé dans la suite
    data.head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>Code du département</th>
          <th>Libellé du départementT1</th>
          <th>VotantsT1</th>
          <th>rHollandeT1</th>
          <th>rSarkozyT1</th>
          <th>rNulT1</th>
          <th>VotantsT2</th>
          <th>rHollandeT2</th>
          <th>rSarkozyT2</th>
          <th>rNulT2</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>1</td>
          <td>AIN</td>
          <td>327812</td>
          <td>0.227459</td>
          <td>0.304090</td>
          <td>0.019685</td>
          <td>326587</td>
          <td>0.427692</td>
          <td>0.572308</td>
          <td>0.059748</td>
        </tr>
        <tr>
          <th>1</th>
          <td>2</td>
          <td>AISNE</td>
          <td>303140</td>
          <td>0.271027</td>
          <td>0.241958</td>
          <td>0.017141</td>
          <td>302076</td>
          <td>0.524020</td>
          <td>0.475980</td>
          <td>0.069704</td>
        </tr>
        <tr>
          <th>2</th>
          <td>3</td>
          <td>ALLIER</td>
          <td>211009</td>
          <td>0.296824</td>
          <td>0.240238</td>
          <td>0.023975</td>
          <td>211132</td>
          <td>0.568861</td>
          <td>0.431139</td>
          <td>0.070686</td>
        </tr>
        <tr>
          <th>3</th>
          <td>4</td>
          <td>ALPES-DE-HAUTE-PROVENCE</td>
          <td>102899</td>
          <td>0.243591</td>
          <td>0.254673</td>
          <td>0.020515</td>
          <td>103581</td>
          <td>0.510594</td>
          <td>0.489406</td>
          <td>0.064095</td>
        </tr>
        <tr>
          <th>4</th>
          <td>5</td>
          <td>HAUTES-ALPES</td>
          <td>88619</td>
          <td>0.244858</td>
          <td>0.261071</td>
          <td>0.020786</td>
          <td>89405</td>
          <td>0.508935</td>
          <td>0.491065</td>
          <td>0.067390</td>
        </tr>
      </tbody>
    </table>
    </div>



.. code:: ipython3

    deps.to_excel("deps.xlsx")
    dict_df["dep1"].to_excel("T1.xlsx")
    dict_df["dep2"].to_excel("T2.xlsx")

De pandas à matplotlib
----------------------

Lorsqu’on construit un graphique avec des données stockées dans un
`DataFrame <http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.html>`__,
on suit généralement le processus suivant :


-  Voir si un graphique correspond dans la page
   `visualisation <http://pandas.pydata.org/pandas-docs/stable/visualization.html>`__
   de pandas
-  Voir la `gallerie <http://matplotlib.org/gallery.html>`__ de
   `matplotlib <http://matplotlib.org/index.html>`__
-  Chercher un exemple de graphique sur un moteur de recherche pour
   tomber sur une page comme celle-ci `Using Python libraries to plot
   two horizontal bar charts sharing same y
   axis <http://stackoverflow.com/questions/27694221/using-python-libraries-to-plot-two-horizontal-bar-charts-sharing-same-y-axis>`__
-  Assembler différentes sources

plot
~~~~

La méthode
`plot <http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.plot.html?highlight=plot#pandas.DataFrame.plot>`__
permet de faire la plupart des graphiques standards (voir
`Plotting <http://pandas.pydata.org/pandas-docs/stable/visualization.html>`__).

.. code:: ipython3

    data.plot(x="Libellé du départementT1", y=["rHollandeT2", "rSarkozyT2"], figsize=(16,5));



.. image:: seance6_graphes_enonce_18_0.png


.. code:: ipython3

    data.plot(x="Libellé du départementT1", y=["rHollandeT2", "rSarkozyT2"],
              figsize=(16,5), kind="bar", stacked=True);



.. image:: seance6_graphes_enonce_19_0.png


.. code:: ipython3

    data.plot(x="rHollandeT1", y="rHollandeT2", figsize=(16,5), 
              kind="scatter", label="s1", title="correlation");



.. image:: seance6_graphes_enonce_20_0.png


superposition
~~~~~~~~~~~~~

La méthode
`plot <http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.plot.html?highlight=plot#pandas.DataFrame.plot>`__
retourne un objet de type
`Axes <http://matplotlib.org/api/axes_api.html>`__. On peut superposer
plusieurs courbes sur le même graphique en s’assurant que la seconde
courbe utilise le même objet.

.. code:: ipython3

    ax = data.plot(x="rHollandeT1", y="rHollandeT2", figsize=(16,5), 
                   kind="scatter", label="H", title="correlation")
    data.plot(x="rSarkozyT1", y="rSarkozyT2", kind="scatter", label="S", ax=ax, c="red");



.. image:: seance6_graphes_enonce_22_0.png


On ajoute une ligne avec la méthode
`Axes.plot <http://matplotlib.org/api/axes_api.html?highlight=plot#matplotlib.axes.Axes.plot>`__
ou du text avec
`text <http://matplotlib.org/api/text_api.html#matplotlib.text.Text>`__
:

.. code:: ipython3

    ax = data.plot(x="rHollandeT1", y="rHollandeT2", figsize=(16,5), 
                   kind="scatter", label="H", title="correlation")
    data.plot(x="rSarkozyT1", y="rSarkozyT2", kind="scatter", label="S", ax=ax, c="red")
    ax.plot([0.2,0.7], [0.2,0.7], "g--")
    ax.text(0.5, 0.5, "rien au dessous", weight="bold", rotation="-30");



.. image:: seance6_graphes_enonce_24_0.png


plusieurs graphes sur la même figure
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

*pandas* crée une
`Figure <http://matplotlib.org/api/pyplot_api.html?highlight=figure#matplotlib.pyplot.figure>`__
de façon implicite avec un seul graphe. Pour créer plusieurs graphes, il
faut créer ce type d’objet en précisant qu’il y aura plusieurs
`Axes <http://matplotlib.org/api/axes_api.html>`__ avec la fonction
`subplots <http://matplotlib.org/api/pyplot_api.html?highlight=figure#matplotlib.pyplot.subplots>`__
et les transmettre à *pandas*. On peut également partager l’axe des X ou
l’axe des Y.

.. code:: ipython3

    import matplotlib.pyplot as plt
    fig, axes = plt.subplots(1, 2, figsize=(16,5), sharey=True)
    data.plot(x="rHollandeT1", y="rHollandeT2", figsize=(16,5), 
              kind="scatter", label="H", ax=axes[0])
    data.plot(x="rSarkozyT1", y="rSarkozyT2", kind="scatter", 
              label="S", ax=axes[1], c="red")
    axes[0].plot([0.2,0.7], [0.2,0.7], "g--")
    axes[1].plot([0.2,0.7], [0.2,0.7], "g--");



.. image:: seance6_graphes_enonce_26_0.png


matplotlib sans pandas
~~~~~~~~~~~~~~~~~~~~~~

On peut se passer de pandas et s’inspirer d’un graphe de la gallerie
pour ajouter des points dépendants du nombre de votants
`scatter_demo <http://matplotlib.org/examples/shapes_and_collections/scatter_demo.html>`__
et ajouter une légende manuellement avec la méthode
`legend <http://matplotlib.org/api/axes_api.html?highlight=legend#matplotlib.axes.Axes.legend>`__.

.. code:: ipython3

    import matplotlib.pyplot as plt
    fig, axes = plt.subplots(1, 1, figsize=(16,5))
    c = axes.scatter(x=data["rHollandeT1"], 
                y=data["rHollandeT2"], 
                s=data["VotantsT1"]/5000, alpha=0.5)
    axes.plot([0.2,0.7], [0.2,0.7], "g--")
    axes.legend( (c,), ("H",) );



.. image:: seance6_graphes_enonce_28_0.png


Pandas et graphes prêts à l’emploi
----------------------------------

histogrammes
~~~~~~~~~~~~

avec
`hist <http://pandas.pydata.org/pandas-docs/stable/generated/pandas.DataFrame.hist.html?highlight=hist#pandas.DataFrame.hist>`__

.. code:: ipython3

    data.hist(figsize=(8, 8));



.. image:: seance6_graphes_enonce_31_0.png


Le paramètre *figsize* permettrait de modifier la taille du graphique.

correlations
~~~~~~~~~~~~

avec
`scatter_matrix <http://pandas.pydata.org/pandas-docs/stable/visualization.html#scatter-matrix-plot>`__

.. code:: ipython3

    from pandas.plotting import scatter_matrix
    scatter_matrix(data, alpha=0.2, figsize=(14, 14), diagonal='kde')
    print("-");


.. parsed-literal::
    -



.. image:: seance6_graphes_enonce_34_1.png


cartes avec cartopy
-------------------

Les coordonnées sur une carte se font avec des `coordonnées
géographiques <https://fr.wikipedia.org/wiki/Coordonn%C3%A9es_g%C3%A9ographiques>`__
: longitude et latitude. La distance entre deux lieux géographiques se
calcule grâce à la `distance de
Haversine <https://en.wikipedia.org/wiki/Haversine_formula>`__. Les
graphes se font avec `cartopy <http://scitools.org.uk/cartopy/>`__.

une carte simple
~~~~~~~~~~~~~~~~

On la choisit centrée sur la France. La carte se dessine avec
`matplotlib <https://matplotlib.org/>`__ auquel
`cartopy <https://scitools.org.uk/cartopy/docs/latest/>`__ ajoute un
système de projection différent. Comme elle accepte un argument ``ax``,
il est possible de changer sa taille ou de la juxtaposer à côté d’un
autre graphe. Le module
`cartopy <https://scitools.org.uk/cartopy/docs/latest/>`__ ne contient
pas toutes les informations sur le territoire français et certaines
`options <https://scitools.org.uk/cartopy/docs/latest/matplotlib/feature_interface.html#cartopy.feature.GSHHSFeature>`__
ne semble pas avoir d’effet.

.. code:: ipython3

    import cartopy.crs as ccrs
    import cartopy.feature as cfeature
    import matplotlib.pyplot as plt
    
    fig = plt.figure(figsize=(7,7))
    ax = fig.add_subplot(1, 1, 1, projection=ccrs.PlateCarree())
    ax.set_extent([-5, 10, 42, 52])
    
    ax.add_feature(cfeature.OCEAN.with_scale('50m'))
    # ax.add_feature(cfeature.COASTLINE)
    # ax.add_feature(cfeature.RIVERS)  # pas d'effet, cartopy ne connaît pas les rivières en France
    # ax.add_feature(cfeature.BORDERS, linestyle=':')
    # cette instruction télécharge un fichier (10m=14Mo, 50m, 110m)
    # il faut la résolution 10m pour la France
    ax.add_feature(cfeature.STATES.with_scale('10m')) 
    ax.set_title('France');



.. image:: seance6_graphes_enonce_37_0.png


La méthode
`with_scale <https://scitools.org.uk/cartopy/docs/latest/matplotlib/feature_interface.html?highlight=with_scale#cartopy.feature.NaturalEarthFeature.with_scale>`__
propose trois résolution 10m, 50m, 110m. La module
`cartopy <https://scitools.org.uk/cartopy/docs/latest/>`__ n’inclut que
la résolution 110m, le reste doit être téléchargé.

exercice 1 : centrer la carte de la France
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


ajouter du texte ou une marque
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Sur une carte, on veut la plupart du temps ajouter du texte. On reprend
le début de ce code qu’on place dans une fonction, puis on place Paris.
On utilise pour cela les fonctions standard de *matplotlib* mais on
convertit les coordonnées géographiques en coordonnées relatives au
graphe (donc dans un repère différent).

.. code:: ipython3

    import cartopy.crs as ccrs
    import cartopy.feature as cfeature
    import matplotlib.pyplot as plt
    
    def carte_france(figsize=(7, 7)):
        fig = plt.figure(figsize=figsize)
        ax = fig.add_subplot(1, 1, 1, projection=ccrs.PlateCarree())
        ax.set_extent([-5, 10, 42, 52])
    
        ax.add_feature(cfeature.OCEAN.with_scale('50m'))
        ax.add_feature(cfeature.RIVERS.with_scale('50m'))
        ax.add_feature(cfeature.BORDERS.with_scale('50m'), linestyle=':')
        ax.set_title('France');
        return ax
    
    carte_france();



.. image:: seance6_graphes_enonce_42_0.png


.. code:: ipython3

    import matplotlib.pyplot as plt
    
    ax = carte_france()
    
    lon = 2.3488000
    lat = 48.853410
    ax.plot([lon], [lat], 'ro', markersize=6)
    ax.text(lon, lat, "Paris");



.. image:: seance6_graphes_enonce_43_0.png


On connaît rarement les coordonnées de chaque ville mais un moteur de
recherche donne rapidement des pistes pour trouver ces données. Il faut
néanmoins s’assurer que le licence autorise ce qu’on l’intention de
faire avec :

`Liste des villes de France en SQL, CSV ou
XML <http://sql.sh/736-base-donnees-villes-francaises>`__

.. code:: ipython3

    from pyensae.datasource import download_data
    download_data("villes_france.csv", url="http://sql.sh/ressources/sql-villes-france/")




.. parsed-literal::
    'villes_france.csv'



.. code:: ipython3

    cols = ["ncommune", "numero_dep", "slug", "nom", "nom_simple", "nom_reel", 
            "nom_soundex", "nom_metaphone", "code_postal",
            "numero_commune", "code_commune", "arrondissement", "canton", 
            "pop2010", "pop1999", "pop2012",
            "densite2010", "surface", "superficie", "dlong", "dlat", "glong", 
            "glat", "slong", "slat", "alt_min", "alt_max"]
    import pandas
    df = pandas.read_csv("villes_france.csv", header=None,low_memory=False, names=cols)

.. code:: ipython3

    df.head()






.. raw:: html

    <div>
    <style scoped>
        .dataframe tbody tr th:only-of-type {
            vertical-align: middle;
        }

        .dataframe tbody tr th {
            vertical-align: top;
        }

        .dataframe thead th {
            text-align: right;
        }
    </style>
    <table border="1" class="dataframe">
      <thead>
        <tr style="text-align: right;">
          <th></th>
          <th>ncommune</th>
          <th>numero_dep</th>
          <th>slug</th>
          <th>nom</th>
          <th>nom_simple</th>
          <th>nom_reel</th>
          <th>nom_soundex</th>
          <th>nom_metaphone</th>
          <th>code_postal</th>
          <th>numero_commune</th>
          <th>...</th>
          <th>surface</th>
          <th>superficie</th>
          <th>dlong</th>
          <th>dlat</th>
          <th>glong</th>
          <th>glat</th>
          <th>slong</th>
          <th>slat</th>
          <th>alt_min</th>
          <th>alt_max</th>
        </tr>
      </thead>
      <tbody>
        <tr>
          <th>0</th>
          <td>1</td>
          <td>01</td>
          <td>ozan</td>
          <td>OZAN</td>
          <td>ozan</td>
          <td>Ozan</td>
          <td>O250</td>
          <td>OSN</td>
          <td>01190</td>
          <td>284</td>
          <td>...</td>
          <td>93</td>
          <td>6.60</td>
          <td>4.91667</td>
          <td>46.3833</td>
          <td>2866.0</td>
          <td>51546.0</td>
          <td>45456.0</td>
          <td>462330.0</td>
          <td>170.0</td>
          <td>205.0</td>
        </tr>
        <tr>
          <th>1</th>
          <td>2</td>
          <td>01</td>
          <td>cormoranche-sur-saone</td>
          <td>CORMORANCHE-SUR-SAONE</td>
          <td>cormoranche sur saone</td>
          <td>Cormoranche-sur-Saône</td>
          <td>C65652625</td>
          <td>KRMRNXSRSN</td>
          <td>01290</td>
          <td>123</td>
          <td>...</td>
          <td>107</td>
          <td>9.85</td>
          <td>4.83333</td>
          <td>46.2333</td>
          <td>2772.0</td>
          <td>51379.0</td>
          <td>44953.0</td>
          <td>461427.0</td>
          <td>168.0</td>
          <td>211.0</td>
        </tr>
        <tr>
          <th>2</th>
          <td>3</td>
          <td>01</td>
          <td>plagne-01</td>
          <td>PLAGNE</td>
          <td>plagne</td>
          <td>Plagne</td>
          <td>P425</td>
          <td>PLKN</td>
          <td>01130</td>
          <td>298</td>
          <td>...</td>
          <td>20</td>
          <td>6.20</td>
          <td>5.73333</td>
          <td>46.1833</td>
          <td>3769.0</td>
          <td>51324.0</td>
          <td>54342.0</td>
          <td>461131.0</td>
          <td>560.0</td>
          <td>922.0</td>
        </tr>
        <tr>
          <th>3</th>
          <td>4</td>
          <td>01</td>
          <td>tossiat</td>
          <td>TOSSIAT</td>
          <td>tossiat</td>
          <td>Tossiat</td>
          <td>T230</td>
          <td>TST</td>
          <td>01250</td>
          <td>422</td>
          <td>...</td>
          <td>138</td>
          <td>10.17</td>
          <td>5.31667</td>
          <td>46.1333</td>
          <td>3309.0</td>
          <td>51268.0</td>
          <td>51854.0</td>
          <td>460828.0</td>
          <td>244.0</td>
          <td>501.0</td>
        </tr>
        <tr>
          <th>4</th>
          <td>5</td>
          <td>01</td>
          <td>pouillat</td>
          <td>POUILLAT</td>
          <td>pouillat</td>
          <td>Pouillat</td>
          <td>P430</td>
          <td>PLT</td>
          <td>01250</td>
          <td>309</td>
          <td>...</td>
          <td>14</td>
          <td>6.23</td>
          <td>5.43333</td>
          <td>46.3333</td>
          <td>3435.0</td>
          <td>51475.0</td>
          <td>52542.0</td>
          <td>461938.0</td>
          <td>333.0</td>
          <td>770.0</td>
        </tr>
      </tbody>
    </table>
    <p>5 rows × 27 columns</p>
    </div>



exercice 2 : placer les plus grandes villes de France sur la carte
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


départements
~~~~~~~~~~~~

Pour dessiner des formes sur une carte, il faut connaître les
coordonnées de ces formes. L’article suivant `Matplotlib Basemap
tutorial 10: Shapefiles Unleached,
continued <http://www.geophysique.be/2013/02/12/matplotlib-basemap-tutorial-10-shapefiles-unleached-continued/>`__
permet de dessiner les départements belges. On va s’en inspirer pour
dessiner les départements français. La première chose à faire est de
récupérer des données géographiques. Une façon simple de les trouver est
d’utiliser un moteur de recherche avec le mot clé **shapefile** inclus
dedans : c’est le format du fichier. *shapefile france* permet d’obtenir
quelques sources. En voici d’autres :


-  `GADM <http://www.gadm.org/>`__ : database of Global Administrative
   Areas
-  `OpenData.gouv
   commune <https://www.data.gouv.fr/fr/datasets/geofla-communes/>`__ :
   base de données sur data.gouv.fr
-  `The National Map Small-Scale
   Collection <http://nationalmap.gov/small_scale/#chpbound>`__ :
   Etats-Unis
-  `ArcGIS <https://developers.arcgis.com/javascript/jsapi/esri.basemaps-amd.html>`__
   : API Javascripts
-  `Natural Earth <http://www.naturalearthdata.com/>`__ : Natural Earth
   is a public domain map dataset available at 1:10m, 1:50m, and 1:110
   million scales. Featuring tightly integrated vector and raster data,
   with Natural Earth you can make a variety of visually pleasing,
   well-crafted maps with cartography or GIS software.
-  `thematicmapping <http://thematicmapping.org/downloads/world_borders.php>`__
   : World Borders Dataset
-  `OpenStreetMap Data Extracts <http://download.geofabrik.de/>`__ :
   OpenStreetMap data
-  `OpenStreetMapData <http://openstreetmapdata.com/>`__ : OpenStreetMap
   data
-  `Shapefile sur
   Wikipedia <http://wiki.openstreetmap.org/wiki/Shapefiles>`__ :
   contient divers liens vers des sources de données

La première chose à vérifier est la licence associées aux données : on
ne peut pas en faire ce qu’on veut. Pour cet exemple, j’ai choisi la
première source de données, GADM. La licence n’est pas précisée
explicitement (on peut trouver *happy to share* sur le site, la page
wikipedia `GADM <https://en.wikipedia.org/wiki/GADM>`__ précise : *GADM
is not freely available for commercial use. The GADM project created the
spatial data for many countries from spatial databases provided by
national governments, NGO, and/or from maps and lists of names available
on the Internet (e.g. from Wikipedia).* C’est le choix que j’avais fait
en 2015 mais l’accès à ces bases a probablement changé car l’accès est
restreint. J’ai donc opté pour les bases accessibles depuis
`data.gouv.fr <https://www.data.gouv.fr/fr/datasets/geofla-departements-30383060/>`__.
Leur seul inconvénient est que les coordonnées sont exprimées dans une
projection de type `Lambert
93 <https://fr.wikipedia.org/wiki/Projection_conique_conforme_de_Lambert>`__.
Cela nécessite une conversion.

.. code:: ipython3

    from pyensae.datasource import download_data
    try:
        download_data("GEOFLA_2-1_DEPARTEMENT_SHP_LAMB93_FXX_2015-12-01.7z",
                  website="https://wxs-telechargement.ign.fr/oikr5jryiph0iwhw36053ptm/telechargement/inspire/" + \
                  "GEOFLA_THEME-DEPARTEMENTS_2015_2$GEOFLA_2-1_DEPARTEMENT_SHP_LAMB93_FXX_2015-12-01/file/")
    except Exception as e:
        # au cas le site n'est pas accessible
        download_data("GEOFLA_2-1_DEPARTEMENT_SHP_LAMB93_FXX_2015-12-01.7z", website="xd")

.. code:: ipython3

    from pyquickhelper.filehelper import un7zip_files
    try:
        un7zip_files("GEOFLA_2-1_DEPARTEMENT_SHP_LAMB93_FXX_2015-12-01.7z", where_to="shapefiles")
        departements = 'shapefiles/GEOFLA_2-1_DEPARTEMENT_SHP_LAMB93_FXX_2015-12-01/GEOFLA/1_DONNEES_LIVRAISON_2015/' + \
                       'GEOFLA_2-1_SHP_LAMB93_FR-ED152/DEPARTEMENT/DEPARTEMENT.shp'
    except FileNotFoundError as e:
        # Il est possible que cette instruction ne fonctionne pas.
        # Dans ce cas, on prendra une copie de ce fichier.
        import warnings
        warnings.warn("Plan B parce que " + str(e))
        download_data("DEPARTEMENT.zip")
        departements = "DEPARTEMENT.shp"
        
    import os
    if not os.path.exists(departements):
        raise FileNotFoundError("Impossible de trouver '{0}'".format(departements))

La license accompagne les données : *ce produit est téléchargeable et
utilisable gratuitement sous
licence*\ `Etalab <https://www.etalab.gouv.fr/licence-ouverte-open-licence>`__.
Pour un usage commercial, il faut faire attentation à la license
associée aux données. Le seul inconvénient des données *GEOFLA* est que
certaines sont données dans le système de coordonnées `Lambert
93 <https://fr.wikipedia.org/wiki/Projection_conique_conforme_de_Lambert>`__
(voir aussi `Cartographie avec
R <https://www.sylvaindurand.org/cartographie-avec-R/>`__).

.. code:: ipython3

    shp = departements
    import shapefile
    r = shapefile.Reader(shp)
    shapes = r.shapes()
    records = r.records()
    len(shapes), len(records)




.. parsed-literal::
    (96, 96)



.. code:: ipython3

    r.bbox




.. parsed-literal::
    [99217.1, 6049646.300000001, 1242417.2, 7110480.100000001]



On regarde une zone en particulier mais on réduit la quantité de données
affichées :

.. code:: ipython3

    d = shapes[0].__dict__.copy()
    d["points"] = d["points"][:10]
    d




.. parsed-literal::
    {'shapeType': 5,
     'points': [(701742.0, 6751181.100000001),
      (701651.9, 6751166.9),
      (701552.0, 6751162.7),
      (700833.7000000001, 6751313.7),
      (700669.4, 6751380.0),
      (700475.4, 6751476.600000001),
      (700400.7000000001, 6751517.2),
      (700098.3, 6751789.600000001),
      (699993.8, 6751845.4),
      (699874.1000000001, 6751876.4)],
     'parts': [0],
     'bbox': [688654.4, 6690595.300000001, 800332.3, 6811114.5]}



350 départements, sûr ?

.. code:: ipython3

    records[0], records[1]




.. parsed-literal::
    (Record #0: ['DEPARTEM0000000000000004', '89', 'YONNE', '024', 'AUXERRE', 742447, 6744261, 748211, 6750855, '27', 'BOURGOGNE-FRANCHE-COMTE'],
     Record #1: ['DEPARTEM0000000000000028', '69', 'RHONE', '381', 'LYON', 842221, 6520526, 832095, 6530600, '84', 'AUVERGNE-RHONE-ALPES'])



.. code:: ipython3

    len(set([r[6] for r in records]))




.. parsed-literal::
    96



Puis je récupère le code final (toujours à `Matplotlib Basemap tutorial
10: Shapefiles Unleached,
continued <http://www.geophysique.be/2013/02/12/matplotlib-basemap-tutorial-10-shapefiles-unleached-continued/>`__)
en l’adaptant pour la France. Petite astuce, on utilie la fonction
`lambert932WGPS <https://github.com/sdpython/ensae_teaching_cs/blob/master/src/ensae_teaching_cs/helpers/geo_helper.py>`__
du module
`ensae_teaching_cs <http://www.xavierdupre.fr/app/ensae_teaching_cs/>`__.
On recopie le code ici :

.. code:: ipython3

    import math
    
    
    def lambert932WGPS(lambertE, lambertN):
    
        class constantes:
            GRS80E = 0.081819191042816
            LONG_0 = 3
            XS = 700000
            YS = 12655612.0499
            n = 0.7256077650532670
            C = 11754255.4261
    
        delX = lambertE - constantes.XS
        delY = lambertN - constantes.YS
        gamma = math.atan(-delX / delY)
        R = math.sqrt(delX * delX + delY * delY)
        latiso = math.log(constantes.C / R) / constantes.n
        sinPhiit0 = math.tanh(latiso + constantes.GRS80E * math.atanh(constantes.GRS80E * math.sin(1)))
        sinPhiit1 = math.tanh(latiso + constantes.GRS80E * math.atanh(constantes.GRS80E * sinPhiit0))
        sinPhiit2 = math.tanh(latiso + constantes.GRS80E * math.atanh(constantes.GRS80E * sinPhiit1))
        sinPhiit3 = math.tanh(latiso + constantes.GRS80E * math.atanh(constantes.GRS80E * sinPhiit2))
        sinPhiit4 = math.tanh(latiso + constantes.GRS80E * math.atanh(constantes.GRS80E * sinPhiit3))
        sinPhiit5 = math.tanh(latiso + constantes.GRS80E * math.atanh(constantes.GRS80E * sinPhiit4))
        sinPhiit6 = math.tanh(latiso + constantes.GRS80E * math.atanh(constantes.GRS80E * sinPhiit5))
    
        longRad = math.asin(sinPhiit6)
        latRad = gamma / constantes.n + constantes.LONG_0 / 180 * math.pi
    
        longitude = latRad / math.pi * 180
        latitude = longRad / math.pi * 180
    
        return longitude, latitude
    
    lambert932WGPS(99217.1, 6049646.300000001), lambert932WGPS(1242417.2, 7110480.100000001)




.. parsed-literal::
    ((-4.1615802638173065, 41.303505287589545),
     (10.699505053975292, 50.85243395553585))



.. code:: ipython3

    import cartopy.crs as ccrs
    import matplotlib.pyplot as plt
    ax = carte_france((8,8))
    
    from matplotlib.collections import LineCollection
    import shapefile
    import geopandas
    from shapely.geometry import Polygon
    from shapely.ops import cascaded_union, unary_union
     
    shp = departements
    r = shapefile.Reader(shp)
    shapes = r.shapes()
    records = r.records()
    
    polys = []
    for i, (record, shape) in enumerate(zip(records, shapes)):
        # les coordonnées sont en Lambert 93
        if i == 0:
            print(record, shape.parts)
        geo_points = [lambert932WGPS(x,y) for x, y in shape.points]
        if len(shape.parts) == 1:
            # Un seul polygone
            poly = Polygon(geo_points)
        else:
            # Il faut les fusionner.
            ind = list(shape.parts) + [len(shape.points)]
            pols = [Polygon(geo_points[ind[i]:ind[i+1]]) for i in range(0, len(shape.parts))]
            try:
                poly = unary_union(pols)
            except Exception as e:
                print("Cannot merge: ", record)
                print([_.length for _ in pols], ind)
                poly = Polygon(geo_points)
        polys.append(poly)
        
    data = geopandas.GeoDataFrame(geometry=polys)
    # cmap -> voir https://matplotlib.org/users/colormaps.html
    data.plot(ax=ax, cmap='tab20', edgecolor='black');
    # Ou pour définir des couleurs spécifiques.
    # geopandas.plotting.plot_polygon_collection(ax, data['geometry'], data['colors'], values=None)


.. parsed-literal::
    Record #0: ['DEPARTEM0000000000000004', '89', 'YONNE', '024', 'AUXERRE', 742447, 6744261, 748211, 6750855, '27', 'BOURGOGNE-FRANCHE-COMTE'] [0]



.. image:: seance6_graphes_enonce_63_1.png


exercice 3 : résultats des élections par départements
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Ce n’est pas toujours évident !


seaborn
-------

`seaborn <http://stanford.edu/~mwaskom/software/seaborn/>`__ propose des
graphiques inéressants pour un statisticien. La
`gallerie <http://stanford.edu/~mwaskom/software/seaborn/examples/index.html>`__
en donne un bon aperçu. On retravaille peu les graphiques. Le code
suivant montre les corrélations entre variables
`pairplot <http://stanford.edu/~mwaskom/software/seaborn/examples/scatterplot_matrix.html>`__.

**warning** les warnings sont la plupart du temps dûs au fait que
*seaborn* a été testé sur une version antérieure d’une de ses
dépendances comme *matplotlib* et qu’il n’est pas encore à jour pour
tenir compte des derniers développement.

.. code:: ipython3

    import seaborn
    seaborn.pairplot(data_elections);



.. image:: seance6_graphes_enonce_67_0.png


Celui-ci est aussi intéressant :
`clustermap <http://stanford.edu/~mwaskom/software/seaborn/examples/structured_heatmap.html>`__
pour étudier les corrélations.

.. code:: ipython3

    import seaborn
    seaborn.set(font="monospace")
    
    cmap = seaborn.diverging_palette(h_neg=210, h_pos=350, s=90, l=30, as_cmap=True)
    
    seaborn.clustermap(data_elections.corr(), linewidths=.5, figsize=(13, 13), cmap=cmap);



.. image:: seance6_graphes_enonce_69_0.png


bokeh
-----

`bokeh <http://bokeh.pydata.org/en/latest/>`__ propose des graphiques en
javascript. La
`gallerie <http://bokeh.pydata.org/en/latest/docs/gallery.html>`__ est
moins fournie que celle de matplotib. Le principale avantage de bokeh
est de proposer gaphiques zoomables (interactifs).

initialisation
~~~~~~~~~~~~~~

La première étape est de signifier que la sortie se fera dans un
notebook.

.. code:: ipython3

    from bokeh.plotting import output_notebook
    output_notebook()





.. raw:: html


    <div class="bk-root">
        <a href="https://bokeh.org" target="_blank" class="bk-logo bk-logo-small bk-logo-notebook"></a>
        <span id="1002">Loading BokehJS ...</span>
    </div>




premier graphe
~~~~~~~~~~~~~~

On utilise bokeh pour un simple graphique XY. on peut choisir les
différentes options interactives, zoom … Voir
`tools <http://bokeh.pydata.org/en/latest/docs/user_guide/tools.html#specifying-tools>`__.

.. code:: ipython3

    from bokeh.plotting import figure, show
    
    p = figure(title = "élections")
    p.title.text = "élections"
    p.circle(data_elections["rHollandeT1"], data_elections["rHollandeT2"], color="red", fill_alpha=0.2, size=10, legend="H")
    p.circle(data_elections["rSarkozyT1"], data_elections["rSarkozyT2"], color="blue", fill_alpha=0.2, size=10, legend="S")
    p.line([0.2,0.7], [0.2,0.7], line_color="green", line_dash="dashed")
    p.xaxis.axis_label = "tour 1"
    p.yaxis.axis_label = "tour 2"
    show(p)


.. parsed-literal::
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead





.. raw:: html







    <div class="bk-root" id="ba2bd55c-fd17-44bf-a58b-56914634eaec" data-root-id="1003"></div>





ajouter du texte
~~~~~~~~~~~~~~~~

Les différents éléments qu’on peut ajouter au graphe s’appelle des
`Glyphes <http://bokeh.pydata.org/en/latest/docs/reference/bokehjs.html>`__.
La documentation manque parfois de précision. Pour la fonction
`text <http://bokeh.pydata.org/en/latest/docs/reference/bokehjs.html#text>`__,
il faut préciser le texte à afficher sous forme de liste.

.. code:: ipython3

    from bokeh.plotting import figure, show
    
    p = figure(title = "élections")
    p.title.text = "élections"
    p.circle(data_elections["rHollandeT1"], data_elections["rHollandeT2"], color="red", fill_alpha=0.2, size=10, legend="H")
    p.circle(data_elections["rSarkozyT1"], data_elections["rSarkozyT2"], color="blue", fill_alpha=0.2, size=10, legend="S")
    p.line([0.2,0.7], [0.2,0.7], line_color="green", line_dash="dashed")
    p.xaxis.axis_label = "tour 1"
    p.yaxis.axis_label = "tour 2"
    
    
    def display_text(p, row):
        p.text(x=row["rHollandeT1"], y=row["rHollandeT2"], text=[row["Libellé du départementT1"]],
              text_font_size="8pt", color="black", text_align="left", text_baseline="middle", angle=0)
        return row["Libellé du départementT1"]
    
    data_elections.apply(lambda row: display_text(p, row), axis=1)
    
    show(p)


.. parsed-literal::
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead





.. raw:: html







    <div class="bk-root" id="f55d78fb-bd81-4f9a-be58-a41875fab5b1" data-root-id="1160"></div>





composition de graphes
~~~~~~~~~~~~~~~~~~~~~~

Voir `linked
brushing <http://bokeh.pydata.org/en/latest/docs/user_guide/interaction.html#linked-brushing>`__.

.. code:: ipython3

    from bokeh.plotting import figure, show, gridplot
    size = 400
    
    ph = figure(title = "élections", width=size, height=size)
    ph.title.text = "élections"
    ph.circle(data_elections["rHollandeT1"], data_elections["rHollandeT2"], color="red", fill_alpha=0.2, size=10, legend="H")
    ph.line([0.2,0.7], [0.2,0.7], line_color="green", line_dash="dashed")
    ph.xaxis.axis_label = "tour 1"
    ph.yaxis.axis_label = "tour 2"
    
    ps = figure(title = "élections", width=size, height=size)
    ps.title.text = "élections"
    ps.circle(data_elections["rSarkozyT1"], data_elections["rSarkozyT2"], color="blue", fill_alpha=0.2, size=10, legend="S")
    ps.line([0.2,0.7], [0.2,0.7], line_color="green", line_dash="dashed")
    ps.xaxis.axis_label = "tour 1"
    ps.yaxis.axis_label = "tour 2"
    
    p = gridplot([[ph, ps]])
    
    show(p)


.. parsed-literal::
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead





.. raw:: html







    <div class="bk-root" id="0fc911a0-3a84-483f-97ec-94236880a383" data-root-id="3289"></div>





interactions avec bokeh
~~~~~~~~~~~~~~~~~~~~~~~

Tout est expliqué dans la page
`interaction <http://bokeh.pydata.org/en/latest/docs/user_guide/interaction.html>`__.

.. code:: ipython3

    from bokeh.models.widgets import Panel, Tabs
    from bokeh.io import show
    from bokeh.plotting import figure
    
    ph = figure(title="élections")
    ph.circle(data_elections["rHollandeT1"], data_elections["rHollandeT2"], color="red", fill_alpha=0.2, size=10, legend="H")
    ph.line([0.2,0.7], [0.2,0.7], line_color="green", line_dash="dashed")
    ph.xaxis.axis_label = "tour 1"
    ph.yaxis.axis_label = "tour 2"
    
    ps = figure(title = "élections")
    ps.title.text = "élections"
    ps.circle(data_elections["rSarkozyT1"], data_elections["rSarkozyT2"], color="blue", fill_alpha=0.2, size=10, legend="S")
    ps.line([0.2,0.7], [0.2,0.7], line_color="green", line_dash="dashed")
    ps.xaxis.axis_label = "tour 1"
    ps.yaxis.axis_label = "tour 2"
    
    tab1 = Panel(child=ph, title="Hollande")
    tab2 = Panel(child=ps, title="Sarkozy")
    
    tabs = Tabs(tabs=[ tab1, tab2 ])
    
    show(tabs)


.. parsed-literal::
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead
    BokehDeprecationWarning: 'legend' keyword is deprecated, use explicit 'legend_label', 'legend_field', or 'legend_group' keywords instead





.. raw:: html







    <div class="bk-root" id="7efdbfe0-da5f-4808-b7ec-8cfd7d703870" data-root-id="3788"></div>





Si on connaît un peu de javascript, on peut créer des graphes qui
peuvent interagir avec la souris à n’importe quel endroit du graphe.

interaction avec matplotlib (ou bokeh)
--------------------------------------

dropdown
~~~~~~~~

Le module `interact <https://github.com/ipython/ipywidgets>`__ permet de
changer le contenu d’une cellule en fonction d’un bouton, d’une barre de
défilement… Le code de cette fonction est sur
`github <https://github.com/ipython/ipywidgets/blob/master/ipywidgets/widgets/interaction.py>`__.
La liste des widget possible est dans le notebook
`WidgetList <https://github.com/ipython/ipywidgets/blob/477cb8046e3217b134762f53c66816c45d688a20/examples/Widget%20List.ipynb>`__

.. code:: ipython3

    import matplotlib.pyplot as plt
    from IPython.html.widgets import interact, Dropdown
    
    def plot(cand):
        
        fig, axes = plt.subplots(1, 1, figsize=(14,5), sharey=True)
        if cand=="Hollande":
            data_elections.plot(x="rHollandeT1", y="rHollandeT2", kind="scatter", label="H", ax=axes)
        else:
            data_elections.plot(x="rSarkozyT1", y="rSarkozyT2", kind="scatter", label="S", ax=axes, c="red")
        axes.plot([0.2,0.7], [0.2,0.7], "g--")
        return axes
    
    cand = Dropdown(options=['Hollande', 'Sarkozy'], value='Hollande', description='candidat')
    
    interact(plot, cand=cand)
    print("")


.. parsed-literal::
    C:\Python395_x64\lib\site-packages\IPython\html.py:12: ShimWarning: The `IPython.html` package has been deprecated since IPython 4.0. You should import from `notebook` instead. `IPython.html.widgets` has moved to `ipywidgets`.
      warn("The `IPython.html` package has been deprecated since IPython 4.0. "



.. parsed-literal::

    interactive(children=(Dropdown(description='candidat', options=('Hollande', 'Sarkozy'), value='Hollande'), Out…


.. parsed-literal::

    



.. image:: seance6_graphes_enonce_86_3.png



.. image:: seance6_graphes_enonce_86_4.png


Ces interactions ne sont pas limitées à matplotlib. Tout type de sortie
peut dépendre d’un *widget*.

exercice 4 : même code, widget différent
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


autres options
--------------

Il existe un grande nombre de modules permettant de dessiner. Les plus
récents utilisent le javascript.

-  static

   -  `ggplot <http://ggplot.yhathq.com/>`__ : aspect similare à
      matplotlib version ggplot mais la syntaxe est différente

-  Javascript

   -  `pygal <http://www.pygal.org/en/latest/>`__ : voir exemple plus
      bas, aspect réussi, le module prévoit une extension pour les
      `carte <http://www.pygal.org/en/latest/documentation/types/maps/pygal_maps_fr.html>`__,
      le résultat nécessite l’ajout de quelques scripts Javascript
   -  `mpld3 <http://mpld3.github.io/>`__ : créer un graphe avec la
      syntaxe Javascript et le transformer en Javascript
   -  `folium <https://github.com/python-visualization/folium>`__ :
      Javascript + `OpenStreetMap <https://www.openstreetmap.org/>`__

.. code:: ipython3

    html_pygal = """
    <script type="text/javascript" src="http://kozea.github.com/pygal.js/javascripts/svg.jquery.js"></script>
    <script type="text/javascript" src="http://kozea.github.com/pygal.js/javascripts/pygal-tooltips.js"></script>
    {pygal_render}
    """
    
    from IPython.core.display import HTML
    import pygal
    xy_chart = pygal.XY(stroke=False)
    xy_chart.title = 'Elections 2012'
    xy = list(zip(data_elections["rHollandeT1"], data_elections["rHollandeT2"]))
    xy_chart.add('H', xy)
    HTML(html_pygal.format(pygal_render=xy_chart.render().decode()))






.. raw:: html


    <script type="text/javascript" src="http://kozea.github.com/pygal.js/javascripts/svg.jquery.js"></script>
    <script type="text/javascript" src="http://kozea.github.com/pygal.js/javascripts/pygal-tooltips.js"></script>
    <?xml version='1.0' encoding='utf-8'?>
    <svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" id="chart-51d436ea-86a6-4a15-b946-ae91f9d60952" class="pygal-chart" viewBox="0 0 800 600"><!--Generated with pygal 3.0.0 (lxml) ©Kozea 2012-2016 on 2021-12-19--><!--http://pygal.org--><!--http://github.com/Kozea/pygal--><defs><style type="text/css">#chart-51d436ea-86a6-4a15-b946-ae91f9d60952{-webkit-user-select:none;-webkit-font-smoothing:antialiased;font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .title{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:16px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .legends .legend text{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:14px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis text{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:10px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis text.major{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:10px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .text-overlay text.value{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:16px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .text-overlay text.label{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:10px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:14px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 text.no_data{font-family:Consolas,"Liberation Mono",Menlo,Courier,monospace;font-size:64px}
    #chart-51d436ea-86a6-4a15-b946-ae91f9d60952{background-color:rgba(249,249,249,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 path,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 rect,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 circle{-webkit-transition:150ms;-moz-transition:150ms;transition:150ms}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .graph &gt; .background{fill:rgba(249,249,249,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .plot &gt; .background{fill:rgba(255,255,255,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .graph{fill:rgba(0,0,0,.87)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 text.no_data{fill:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .title{fill:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .legends .legend text{fill:rgba(0,0,0,.87)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .legends .legend:hover text{fill:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis .line{stroke:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis .guide.line{stroke:rgba(0,0,0,.54)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis .major.line{stroke:rgba(0,0,0,.87)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis text.major{fill:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.y .guides:hover .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .line-graph .axis.x .guides:hover .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .stackedline-graph .axis.x .guides:hover .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .xy-graph .axis.x .guides:hover .guide.line{stroke:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis .guides:hover text{fill:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .reactive{fill-opacity:.7;stroke-opacity:.8;stroke-width:1}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .ci{stroke:rgba(0,0,0,.87)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .reactive.active,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .active .reactive{fill-opacity:.8;stroke-opacity:.9;stroke-width:4}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .ci .reactive.active{stroke-width:1.5}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .series text{fill:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip rect{fill:rgba(255,255,255,1);stroke:rgba(0,0,0,1);-webkit-transition:opacity 150ms;-moz-transition:opacity 150ms;transition:opacity 150ms}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip .label{fill:rgba(0,0,0,.87)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip .label{fill:rgba(0,0,0,.87)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip .legend{font-size:.8em;fill:rgba(0,0,0,.54)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip .x_label{font-size:.6em;fill:rgba(0,0,0,1)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip .xlink{font-size:.5em;text-decoration:underline}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip .value{font-size:1.5em}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .bound{font-size:.5em}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .max-value{font-size:.75em;fill:rgba(0,0,0,.54)}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .map-element{fill:rgba(255,255,255,1);stroke:rgba(0,0,0,.54) !important}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .map-element .reactive{fill-opacity:inherit;stroke-opacity:inherit}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .color-0,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .color-0 a:visited{stroke:#F44336;fill:#F44336}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .text-overlay .color-0 text{fill:black}
    #chart-51d436ea-86a6-4a15-b946-ae91f9d60952 text.no_data{text-anchor:middle}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .guide.line{fill:none}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .centered{text-anchor:middle}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .title{text-anchor:middle}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .legends .legend text{fill-opacity:1}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.x text{text-anchor:middle}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.x:not(.web) text[transform]{text-anchor:start}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.x:not(.web) text[transform].backwards{text-anchor:end}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.y text{text-anchor:end}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.y text[transform].backwards{text-anchor:start}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.y2 text{text-anchor:start}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.y2 text[transform].backwards{text-anchor:end}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis .guide.line{stroke-dasharray:4,4;stroke:black}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis .major.guide.line{stroke-dasharray:6,6;stroke:black}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .horizontal .axis.y .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .horizontal .axis.y2 .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .vertical .axis.x .guide.line{opacity:0}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .horizontal .axis.always_show .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .vertical .axis.always_show .guide.line{opacity:1 !important}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.y .guides:hover .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.y2 .guides:hover .guide.line,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis.x .guides:hover .guide.line{opacity:1}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .axis .guides:hover text{opacity:1}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .nofill{fill:none}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .subtle-fill{fill-opacity:.2}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .dot{stroke-width:1px;fill-opacity:1;stroke-opacity:1}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .dot.active{stroke-width:5px}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .dot.negative{fill:transparent}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 text,#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 tspan{stroke:none !important}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .series text.active{opacity:1}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip rect{fill-opacity:.95;stroke-width:.5}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .tooltip text{fill-opacity:1}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .showable{visibility:hidden}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .showable.shown{visibility:visible}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .gauge-background{fill:rgba(229,229,229,1);stroke:none}#chart-51d436ea-86a6-4a15-b946-ae91f9d60952 .bg-lines{stroke:rgba(249,249,249,1);stroke-width:2px}</style><script type="text/javascript">window.pygal = window.pygal || {};window.pygal.config = window.pygal.config || {};window.pygal.config['51d436ea-86a6-4a15-b946-ae91f9d60952'] = {"allow_interruptions": false, "box_mode": "extremes", "classes": ["pygal-chart"], "css": ["file://style.css", "file://graph.css"], "defs": [], "disable_xml_declaration": false, "dots_size": 2.5, "dynamic_print_values": false, "explicit_size": false, "fill": false, "force_uri_protocol": "https", "formatter": null, "half_pie": false, "height": 600, "include_x_axis": false, "inner_radius": 0, "interpolate": null, "interpolation_parameters": {}, "interpolation_precision": 250, "inverse_y_axis": false, "js": ["//kozea.github.io/pygal.js/2.0.x/pygal-tooltips.min.js"], "legend_at_bottom": false, "legend_at_bottom_columns": null, "legend_box_size": 12, "logarithmic": false, "margin": 20, "margin_bottom": null, "margin_left": null, "margin_right": null, "margin_top": null, "max_scale": 16, "min_scale": 4, "missing_value_fill_truncation": "x", "no_data_text": "No data", "no_prefix": false, "order_min": null, "pretty_print": false, "print_labels": false, "print_values": false, "print_values_position": "center", "print_zeroes": true, "range": null, "rounded_bars": null, "secondary_range": null, "show_dots": true, "show_legend": true, "show_minor_x_labels": true, "show_minor_y_labels": true, "show_only_major_dots": false, "show_x_guides": false, "show_x_labels": true, "show_y_guides": true, "show_y_labels": true, "spacing": 10, "stack_from_top": false, "strict": false, "stroke": false, "stroke_style": null, "style": {"background": "rgba(249, 249, 249, 1)", "ci_colors": [], "colors": ["#F44336", "#3F51B5", "#009688", "#FFC107", "#FF5722", "#9C27B0", "#03A9F4", "#8BC34A", "#FF9800", "#E91E63", "#2196F3", "#4CAF50", "#FFEB3B", "#673AB7", "#00BCD4", "#CDDC39", "#9E9E9E", "#607D8B"], "dot_opacity": "1", "font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "foreground": "rgba(0, 0, 0, .87)", "foreground_strong": "rgba(0, 0, 0, 1)", "foreground_subtle": "rgba(0, 0, 0, .54)", "guide_stroke_color": "black", "guide_stroke_dasharray": "4,4", "label_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "label_font_size": 10, "legend_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "legend_font_size": 14, "major_guide_stroke_color": "black", "major_guide_stroke_dasharray": "6,6", "major_label_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "major_label_font_size": 10, "no_data_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "no_data_font_size": 64, "opacity": ".7", "opacity_hover": ".8", "plot_background": "rgba(255, 255, 255, 1)", "stroke_opacity": ".8", "stroke_opacity_hover": ".9", "stroke_width": "1", "stroke_width_hover": "4", "title_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "title_font_size": 16, "tooltip_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "tooltip_font_size": 14, "transition": "150ms", "value_background": "rgba(229, 229, 229, 1)", "value_colors": [], "value_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "value_font_size": 16, "value_label_font_family": "Consolas, \"Liberation Mono\", Menlo, Courier, monospace", "value_label_font_size": 10}, "title": "Elections 2012", "tooltip_border_radius": 0, "tooltip_fancy_mode": true, "truncate_label": null, "truncate_legend": null, "width": 800, "x_label_rotation": 0, "x_labels": null, "x_labels_major": null, "x_labels_major_count": null, "x_labels_major_every": null, "x_title": null, "xrange": null, "y_label_rotation": 0, "y_labels": null, "y_labels_major": null, "y_labels_major_count": null, "y_labels_major_every": null, "y_title": null, "zero": 0, "legends": ["H"]}</script><script type="text/javascript" xlink:href="https://kozea.github.io/pygal.js/2.0.x/pygal-tooltips.min.js"/></defs><title>Elections 2012</title><g class="graph xy-graph vertical"><rect x="0" y="0" width="800" height="600" class="background"/><g transform="translate(84, 46)" class="plot"><rect x="0" y="0" width="695.6" height="514" class="background"/><g class="axis y always_show"><path d="M0.000000 514.000000 h695.600000" class="line"/><g class="guides"><path d="M0.000000 499.839829 h695.600000" class="guide line"/><text x="-5" y="503.33982892854016" class="">0.36</text><title>0.36</title></g><g class="guides"><path d="M0.000000 445.304576 h695.600000" class="guide line"/><text x="-5" y="448.804576224942" class="">0.4</text><title>0.4</title></g><g class="guides"><path d="M0.000000 390.769324 h695.600000" class="guide line"/><text x="-5" y="394.26932352134395" class="">0.44</text><title>0.44</title></g><g class="guides"><path d="M0.000000 336.234071 h695.600000" class="guide line"/><text x="-5" y="339.7340708177459" class="">0.48</text><title>0.48</title></g><g class="guides"><path d="M0.000000 281.698818 h695.600000" class="guide line"/><text x="-5" y="285.1988181141478" class="">0.52</text><title>0.52</title></g><g class="guides"><path d="M0.000000 227.163565 h695.600000" class="guide line"/><text x="-5" y="230.66356541054967" class="">0.56</text><title>0.56</title></g><g class="guides"><path d="M0.000000 172.628313 h695.600000" class="guide line"/><text x="-5" y="176.1283127069517" class="">0.6</text><title>0.6</title></g><g class="guides"><path d="M0.000000 118.093060 h695.600000" class="guide line"/><text x="-5" y="121.59306000335357" class="">0.64</text><title>0.64</title></g><g class="guides"><path d="M0.000000 63.557807 h695.600000" class="guide line"/><text x="-5" y="67.05780729975538" class="">0.68</text><title>0.68</title></g></g><g class="axis x"><path d="M0.000000 0.000000 v514.000000" class="line"/><g class="guides"><path d="M32.868848 0.000000 v514.000000" class="guide line"/><text x="32.868848074862406" y="529.0" class="">0.2</text><title>0.2</title></g><g class="guides"><path d="M103.076665 0.000000 v514.000000" class="guide line"/><text x="103.07666474528646" y="529.0" class="">0.24</text><title>0.24</title></g><g class="guides"><path d="M173.284481 0.000000 v514.000000" class="guide line"/><text x="173.28448141571062" y="529.0" class="">0.28</text><title>0.28</title></g><g class="guides"><path d="M243.492298 0.000000 v514.000000" class="guide line"/><text x="243.49229808613472" y="529.0" class="">0.32</text><title>0.32</title></g><g class="guides"><path d="M313.700115 0.000000 v514.000000" class="guide line"/><text x="313.7001147565588" y="529.0" class="">0.36</text><title>0.36</title></g><g class="guides"><path d="M383.907931 0.000000 v514.000000" class="guide line"/><text x="383.90793142698294" y="529.0" class="">0.4</text><title>0.4</title></g><g class="guides"><path d="M454.115748 0.000000 v514.000000" class="guide line"/><text x="454.115748097407" y="529.0" class="">0.44</text><title>0.44</title></g><g class="guides"><path d="M524.323565 0.000000 v514.000000" class="guide line"/><text x="524.323564767831" y="529.0" class="">0.48</text><title>0.48</title></g><g class="guides"><path d="M594.531381 0.000000 v514.000000" class="guide line"/><text x="594.5313814382553" y="529.0" class="">0.52</text><title>0.52</title></g><g class="guides"><path d="M664.739198 0.000000 v514.000000" class="guide line"/><text x="664.7391981086794" y="529.0" class="">0.56</text><title>0.56</title></g></g><g class="series serie-0 color-0"/></g><g class="titles"><text x="400.0" y="26" class="title plot_title">Elections 2012</text></g><g transform="translate(84, 46)" class="plot overlay"><g class="series serie-0 color-0"><g class="dots"><circle cx="81.0647766051942" cy="407.5502346134921" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.22745901: 0.4276916965</desc><desc class="x top">81.0647766051942</desc><desc class="y top">407.5502346134921</desc></g><g class="dots"><circle cx="157.53588768153753" cy="276.21851231260496" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2710274414: 0.5240196427</desc><desc class="x top">157.53588768153753</desc><desc class="y top">276.21851231260496</desc></g><g class="dots"><circle cx="202.81471748793922" cy="215.08319311045756" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.296824472: 0.5688605969</desc><desc class="x ">202.81471748793922</desc><desc class="y ">215.08319311045756</desc></g><g class="dots"><circle cx="109.37870583566756" cy="294.5228326860695" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2435905068: 0.5105939634</desc><desc class="x top">109.37870583566756</desc><desc class="y top">294.5228326860695</desc></g><g class="dots"><circle cx="111.6025410958509" cy="296.7846372987063" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2448575083: 0.5089349964</desc><desc class="x top">111.6025410958509</desc><desc class="y top">296.7846372987063</desc></g><g class="dots"><circle cx="19.081257791223354" cy="504.11538461538464" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.1921446979: 0.356864006</desc><desc class="x top">19.081257791223354</desc><desc class="y top">504.11538461538464</desc></g><g class="dots"><circle cx="137.91228924254577" cy="261.96529915133124" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2598471487: 0.5344739544</desc><desc class="x top">137.91228924254577</desc><desc class="y top">261.96529915133124</desc></g><g class="dots"><circle cx="189.68920646254398" cy="283.20070489762554" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.289346381: 0.5188984103</desc><desc class="x top">189.68920646254398</desc><desc class="y top">283.20070489762554</desc></g><g class="dots"><circle cx="284.9314882292195" cy="108.62275273323422" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3436094453: 0.6469461912</desc><desc class="x ">284.9314882292195</desc><desc class="y ">108.62275273323422</desc></g><g class="dots"><circle cx="81.89332776118067" cy="409.4453475169565" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2279310664: 0.426301687</desc><desc class="x top">81.89332776118067</desc><desc class="y top">409.4453475169565</desc></g><g class="dots"><circle cx="215.51632572983564" cy="223.73808239394953" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3040610498: 0.5625124908</desc><desc class="x ">215.51632572983564</desc><desc class="y ">223.73808239394953</desc></g><g class="dots"><circle cx="198.52735063445377" cy="248.6409399276793" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2943817999: 0.544246979</desc><desc class="x ">198.52735063445377</desc><desc class="y ">248.6409399276793</desc></g><g class="dots"><circle cx="112.02023854046952" cy="347.54217674861457" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2450954861: 0.4717058377</desc><desc class="x top">112.02023854046952</desc><desc class="y top">347.54217674861457</desc></g><g class="dots"><circle cx="196.77284304770458" cy="266.43873401567777" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2933821918: 0.5311928218</desc><desc class="x top">196.77284304770458</desc><desc class="y top">266.43873401567777</desc></g><g class="dots"><circle cx="223.21389256997236" cy="284.4044828368431" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3084466394: 0.5180154747</desc><desc class="x top">223.21389256997236</desc><desc class="y top">284.4044828368431</desc></g><g class="dots"><circle cx="257.4370595761621" cy="188.58020283166167" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3279448484: 0.5882997589</desc><desc class="x ">257.4370595761621</desc><desc class="y ">188.58020283166167</desc></g><g class="dots"><circle cx="181.50348642317812" cy="287.6143385439449" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2846826723: 0.5156611402</desc><desc class="x top">181.50348642317812</desc><desc class="y top">287.6143385439449</desc></g><g class="dots"><circle cx="151.59569164452353" cy="253.82526567041805" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2676430912: 0.5404444289</desc><desc class="x ">151.59569164452353</desc><desc class="y ">253.82526567041805</desc></g><g class="dots"><circle cx="435.99201592959423" cy="106.32972583947594" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.4296742368: 0.6486280588</desc><desc class="x left">435.99201592959423</desc><desc class="y left">106.32972583947594</desc></g><g class="dots"><circle cx="169.5267356040422" cy="330.068788759956" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2778590727: 0.4845220526</desc><desc class="x top">169.5267356040422</desc><desc class="y top">330.068788759956</desc></g><g class="dots"><circle cx="261.440994112384" cy="183.6588110791036" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3302260386: 0.5919094547</desc><desc class="x ">261.440994112384</desc><desc class="y ">183.6588110791036</desc></g><g class="dots"><circle cx="278.8671047493355" cy="158.76526227641853" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3401543408: 0.6101681388</desc><desc class="x ">278.8671047493355</desc><desc class="y ">158.76526227641853</desc></g><g class="dots"><circle cx="245.04883077777768" cy="184.41384162624013" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3208868145: 0.591355662</desc><desc class="x ">245.04883077777768</desc><desc class="y ">184.41384162624013</desc></g><g class="dots"><circle cx="143.14888459607607" cy="335.04010005686723" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2628306315: 0.4808757424</desc><desc class="x top">143.14888459607607</desc><desc class="y top">335.04010005686723</desc></g><g class="dots"><circle cx="121.50392408286245" cy="320.08883765614746" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2504986939: 0.4918420525</desc><desc class="x top">121.50392408286245</desc><desc class="y top">320.08883765614746</desc></g><g class="dots"><circle cx="114.56893759888008" cy="342.4259936361486" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2465475746: 0.4754584072</desc><desc class="x top">114.56893759888008</desc><desc class="y top">342.4259936361486</desc></g><g class="dots"><circle cx="133.06725965853144" cy="356.3157445661086" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2570867555: 0.4652706844</desc><desc class="x top">133.06725965853144</desc><desc class="y top">356.3157445661086</desc></g><g class="dots"><circle cx="273.25457983769303" cy="188.12144762906854" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.336956677: 0.5886362423</desc><desc class="x ">273.25457983769303</desc><desc class="y ">188.12144762906854</desc></g><g class="dots"><circle cx="104.97350150059782" cy="325.31400824897514" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2410806983: 0.4880095439</desc><desc class="x top">104.97350150059782</desc><desc class="y top">325.31400824897514</desc></g><g class="dots"><circle cx="258.0311110732751" cy="189.27553083416848" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3282833016: 0.5877897563</desc><desc class="x ">258.0311110732751</desc><desc class="y ">189.27553083416848</desc></g><g class="dots"><circle cx="241.0613874895372" cy="218.42590940618663" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.31861502: 0.5664088131</desc><desc class="x ">241.0613874895372</desc><desc class="y ">218.42590940618663</desc></g><g class="dots"><circle cx="239.19078266702041" cy="218.8473694619948" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3175492669: 0.5660996845</desc><desc class="x ">239.19078266702041</desc><desc class="y ">218.8473694619948</desc></g><g class="dots"><circle cx="150.30669394784496" cy="291.05038354827286" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2669087013: 0.5131409025</desc><desc class="x top">150.30669394784496</desc><desc class="y top">291.05038354827286</desc></g><g class="dots"><circle cx="239.38999342035038" cy="231.05538051867148" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3176627647: 0.5571454684</desc><desc class="x ">239.38999342035038</desc><desc class="y ">231.05538051867148</desc></g><g class="dots"><circle cx="206.12220342390205" cy="231.79245831651906" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2987088695: 0.5566048435</desc><desc class="x ">206.12220342390205</desc><desc class="y ">231.79245831651906</desc></g><g class="dots"><circle cx="175.05718925933752" cy="292.2361673474395" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2810099775: 0.512271165</desc><desc class="x top">175.05718925933752</desc><desc class="y top">292.2361673474395</desc></g><g class="dots"><circle cx="170.55314311561494" cy="280.0007798641741" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2784438552: 0.521245461</desc><desc class="x top">170.55314311561494</desc><desc class="y top">280.0007798641741</desc></g><g class="dots"><circle cx="113.64339836862091" cy="316.80710996228686" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2460202605: 0.494249103</desc><desc class="x top">113.64339836862091</desc><desc class="y top">316.80710996228686</desc></g><g class="dots"><circle cx="256.70431329313305" cy="213.68229904062736" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3275273756: 0.5698881114</desc><desc class="x ">256.70431329313305</desc><desc class="y ">213.68229904062736</desc></g><g class="dots"><circle cx="120.82210754353578" cy="342.0332979866088" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2501102377: 0.4757464378</desc><desc class="x top">120.82210754353578</desc><desc class="y top">342.0332979866088</desc></g><g class="dots"><circle cx="146.2326575481917" cy="302.13277339911747" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2645875715: 0.5050122963</desc><desc class="x top">146.2326575481917</desc><desc class="y top">302.13277339911747</desc></g><g class="dots"><circle cx="144.55859315948732" cy="290.159094541105" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2636337949: 0.5137946367</desc><desc class="x top">144.55859315948732</desc><desc class="y top">290.159094541105</desc></g><g class="dots"><circle cx="239.22060466449628" cy="222.34125171271074" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3175662576: 0.5635370249</desc><desc class="x ">239.22060466449628</desc><desc class="y ">222.34125171271074</desc></g><g class="dots"><circle cx="128.54448640285398" cy="363.8730617712888" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2545099636: 0.459727615</desc><desc class="x top">128.54448640285398</desc><desc class="y top">363.8730617712888</desc></g><g class="dots"><circle cx="287.60068049768853" cy="146.87074135272184" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3451301832: 0.6188924192</desc><desc class="x ">287.60068049768853</desc><desc class="y ">146.87074135272184</desc></g><g class="dots"><circle cx="154.2085690839581" cy="290.5050216922191" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2691317444: 0.5135409093</desc><desc class="x top">154.2085690839581</desc><desc class="y top">290.5050216922191</desc></g><g class="dots"><circle cx="123.34884538169514" cy="309.6051701753556" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.251549814: 0.4995315135</desc><desc class="x top">123.34884538169514</desc><desc class="y top">309.6051701753556</desc></g><g class="dots"><circle cx="157.5628832139739" cy="324.6081235312291" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2710428218: 0.4885272896</desc><desc class="x top">157.5628832139739</desc><desc class="y top">324.6081235312291</desc></g><g class="dots"><circle cx="156.41005679414104" cy="310.320120964333" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2703860137: 0.4990071182</desc><desc class="x top">156.41005679414104</desc><desc class="y top">310.320120964333</desc></g><g class="dots"><circle cx="103.99342511879745" cy="381.4152129611241" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2405223124: 0.4468609643</desc><desc class="x top">103.99342511879745</desc><desc class="y top">381.4152129611241</desc></g><g class="dots"><circle cx="98.6672083527552" cy="369.2965639051529" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.237487769: 0.4557496361</desc><desc class="x top">98.6672083527552</desc><desc class="y top">369.2965639051529</desc></g><g class="dots"><circle cx="135.82652294865477" cy="350.8474904453368" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2586588102: 0.4692814875</desc><desc class="x top">135.82652294865477</desc><desc class="y top">350.8474904453368</desc></g><g class="dots"><circle cx="171.9212106021703" cy="267.208506378184" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.279223294: 0.5306282164</desc><desc class="x top">171.9212106021703</desc><desc class="y top">267.208506378184</desc></g><g class="dots"><circle cx="92.5262006096088" cy="360.7774612721839" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2339890088: 0.4619981467</desc><desc class="x top">92.5262006096088</desc><desc class="y top">360.7774612721839</desc></g><g class="dots"><circle cx="178.34182285177977" cy="285.3456103165523" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2828813552: 0.5173251854</desc><desc class="x top">178.34182285177977</desc><desc class="y top">285.3456103165523</desc></g><g class="dots"><circle cx="112.3599890058706" cy="356.6366910312105" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2452890545: 0.4650352796</desc><desc class="x top">112.3599890058706</desc><desc class="y top">356.6366910312105</desc></g><g class="dots"><circle cx="254.78556591053663" cy="188.82355075401722" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.326434194: 0.5881212704</desc><desc class="x ">254.78556591053663</desc><desc class="y ">188.82355075401722</desc></g><g class="dots"><circle cx="172.72541971235609" cy="269.70956467920496" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2796814818: 0.5287937639</desc><desc class="x top">172.72541971235609</desc><desc class="y top">269.70956467920496</desc></g><g class="dots"><circle cx="118.94475954890072" cy="345.2923471005022" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2490406428: 0.4733560214</desc><desc class="x top">118.94475954890072</desc><desc class="y top">345.2923471005022</desc></g><g class="dots"><circle cx="107.70360690887996" cy="348.3280296372793" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2426361408: 0.4711294378</desc><desc class="x top">107.70360690887996</desc><desc class="y top">348.3280296372793</desc></g><g class="dots"><circle cx="198.52721181098605" cy="224.65007159868316" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2943817208: 0.5618435736</desc><desc class="x ">198.52721181098605</desc><desc class="y ">224.65007159868316</desc></g><g class="dots"><circle cx="262.60008045796314" cy="166.26184405232027" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3308864131: 0.6046696171</desc><desc class="x ">262.60008045796314</desc><desc class="y ">166.26184405232027</desc></g><g class="dots"><circle cx="207.16626220401236" cy="211.88894872167486" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2993037086: 0.571203481</desc><desc class="x ">207.16626220401236</desc><desc class="y ">211.88894872167486</desc></g><g class="dots"><circle cx="263.84229846350917" cy="139.01345963263958" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3315941508: 0.6246555037</desc><desc class="x ">263.84229846350917</desc><desc class="y ">139.01345963263958</desc></g><g class="dots"><circle cx="137.5332221133715" cy="300.97668938730124" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2596311801: 0.5058602498</desc><desc class="x top">137.5332221133715</desc><desc class="y top">300.97668938730124</desc></g><g class="dots"><circle cx="25.42667476225948" cy="492.2115699751194" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.1957599175: 0.365595103</desc><desc class="x top">25.42667476225948</desc><desc class="y top">492.2115699751194</desc></g><g class="dots"><circle cx="13.376923076923061" cy="490.765680913056" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.1888947266: 0.36665562</desc><desc class="x top">13.376923076923061</desc><desc class="y top">490.765680913056</desc></g><g class="dots"><circle cx="154.13004051252685" cy="336.5382072536602" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2690870038: 0.4797769249</desc><desc class="x top">154.13004051252685</desc><desc class="y top">336.5382072536602</desc></g><g class="dots"><circle cx="144.9078479745138" cy="313.9097675384362" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2638327783: 0.4963742183</desc><desc class="x top">144.9078479745138</desc><desc class="y top">313.9097675384362</desc></g><g class="dots"><circle cx="185.42461997218203" cy="283.65983820449475" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2869166877: 0.5185616496</desc><desc class="x top">185.42461997218203</desc><desc class="y top">283.65983820449475</desc></g><g class="dots"><circle cx="175.5097606016877" cy="272.5696898826602" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2812678242: 0.5266959464</desc><desc class="x top">175.5097606016877</desc><desc class="y top">272.5696898826602</desc></g><g class="dots"><circle cx="96.73351564644082" cy="348.95823100016685" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2363860724: 0.4706672037</desc><desc class="x top">96.73351564644082</desc><desc class="y top">348.95823100016685</desc></g><g class="dots"><circle cx="41.94575135321092" cy="446.70472011450676" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2051714488: 0.3989730357</desc><desc class="x top">41.94575135321092</desc><desc class="y top">446.70472011450676</desc></g><g class="dots"><circle cx="293.0870549611268" cy="232.57017664592274" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3482559745: 0.5560344101</desc><desc class="x ">293.0870549611268</desc><desc class="y ">232.57017664592274</desc></g><g class="dots"><circle cx="197.85748479486088" cy="241.58227911914298" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2940001524: 0.5494242986</desc><desc class="x ">197.85748479486088</desc><desc class="y ">241.58227911914298</desc></g><g class="dots"><circle cx="167.07023329370008" cy="319.1494116691325" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2764595121: 0.4925310938</desc><desc class="x top">167.07023329370008</desc><desc class="y top">319.1494116691325</desc></g><g class="dots"><circle cx="161.2727491567879" cy="367.52680025178734" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2731564701: 0.4570477056</desc><desc class="x top">161.2727491567879</desc><desc class="y top">367.52680025178734</desc></g><g class="dots"><circle cx="266.9424023349598" cy="209.34170378583684" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3333603951: 0.5730718101</desc><desc class="x ">266.9424023349598</desc><desc class="y ">209.34170378583684</desc></g><g class="dots"><circle cx="179.79381258636184" cy="248.8607044411268" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2837086077: 0.5440857882</desc><desc class="x ">179.79381258636184</desc><desc class="y ">248.8607044411268</desc></g><g class="dots"><circle cx="220.4953334740937" cy="233.2385064770641" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3068977754: 0.5555442098</desc><desc class="x ">220.4953334740937</desc><desc class="y ">233.2385064770641</desc></g><g class="dots"><circle cx="166.79390249462145" cy="291.8914169127388" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2763020762: 0.5125240293</desc><desc class="x top">166.79390249462145</desc><desc class="y top">291.8914169127388</desc></g><g class="dots"><circle cx="26.665928510694187" cy="481.2865020011925" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.1964659664: 0.3736083183</desc><desc class="x top">26.665928510694187</desc><desc class="y top">481.2865020011925</desc></g><g class="dots"><circle cx="73.82035989354489" cy="396.6787879343717" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2233315968: 0.4356655821</desc><desc class="x top">73.82035989354489</desc><desc class="y top">396.6787879343717</desc></g><g class="dots"><circle cx="117.32403433260615" cy="385.29307348632767" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2481172555: 0.4440166679</desc><desc class="x top">117.32403433260615</desc><desc class="y top">385.29307348632767</desc></g><g class="dots"><circle cx="243.279408577325" cy="211.2895781135456" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3198787089: 0.5716431017</desc><desc class="x ">243.279408577325</desc><desc class="y ">211.2895781135456</desc></g><g class="dots"><circle cx="312.016588403223" cy="118.1759498639563" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3590408325: 0.6399392027</desc><desc class="x ">312.016588403223</desc><desc class="y ">118.1759498639563</desc></g><g class="dots"><circle cx="115.0272250764751" cy="321.7670889865854" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2468086779: 0.4906111046</desc><desc class="x top">115.0272250764751</desc><desc class="y top">321.7670889865854</desc></g><g class="dots"><circle cx="103.0113219018644" cy="351.4475677366729" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2399627718: 0.4688413485</desc><desc class="x top">103.0113219018644</desc><desc class="y top">351.4475677366729</desc></g><g class="dots"><circle cx="138.27474824133535" cy="301.83159034087583" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2600536551: 0.5052332051</desc><desc class="x top">138.27474824133535</desc><desc class="y top">301.83159034087583</desc></g><g class="dots"><circle cx="215.23782819111605" cy="262.1797727507367" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3039023794: 0.5343166443</desc><desc class="x top">215.23782819111605</desc><desc class="y top">262.1797727507367</desc></g><g class="dots"><circle cx="211.10951538954086" cy="315.9922235925512" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3015503263: 0.4948467982</desc><desc class="x top">211.10951538954086</desc><desc class="y top">315.9922235925512</desc></g><g class="dots"><circle cx="360.80075409648975" cy="100.10885586746161" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3868349831: 0.6531908835</desc><desc class="x left">360.80075409648975</desc><desc class="y left">100.10885586746161</desc></g><g class="dots"><circle cx="259.8794632657387" cy="220.67593931657797" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3293363765: 0.5647584825</desc><desc class="x ">259.8794632657387</desc><desc class="y ">220.67593931657797</desc></g><g class="dots"><circle cx="250.7052990460707" cy="255.6825676520387" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3241095145: 0.5390821527</desc><desc class="x ">250.7052990460707</desc><desc class="y ">255.6825676520387</desc></g><g class="dots"><circle cx="178.82772604226005" cy="350.5543493445168" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2831581923: 0.4694964979</desc><desc class="x top">178.82772604226005</desc><desc class="y top">350.5543493445168</desc></g><g class="dots"><circle cx="682.223076923077" cy="9.884615384615358" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.5699612149: 0.7193677038</desc><desc class="x left">682.223076923077</desc><desc class="y left">9.884615384615358</desc></g><g class="dots"><circle cx="594.2145894850012" cy="57.65217867127899" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.5198195119: 0.6843316045</desc><desc class="x left">594.2145894850012</desc><desc class="y left">57.65217867127899</desc></g><g class="dots"><circle cx="429.80048346042724" cy="144.71348064697037" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.4261466909: 0.6204747063</desc><desc class="x left">429.80048346042724</desc><desc class="y left">144.71348064697037</desc></g><g class="dots"><circle cx="617.1160511951529" cy="16.02591612969661" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.5328673249: 0.7148632408</desc><desc class="x left">617.1160511951529</desc><desc class="y left">16.02591612969661</desc></g><g class="dots"><circle cx="274.2332349178825" cy="100.28563054911751" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3375142531: 0.6530612245</desc><desc class="x ">274.2332349178825</desc><desc class="y ">100.28563054911751</desc></g><g class="dots"><circle cx="151.81922587698762" cy="437.519571229854" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2677704469: 0.4057100716</desc><desc class="x top">151.81922587698762</desc><desc class="y top">437.519571229854</desc></g><g class="dots"><circle cx="323.3641037504895" cy="321.85117806131564" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.3655059334: 0.4905494278</desc><desc class="x top">323.3641037504895</desc><desc class="y top">321.85117806131564</desc></g><g class="dots"><circle cx="529.1560366191901" cy="226.2853939606025" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.4827532386: 0.5606441129</desc><desc class="x left">529.1560366191901</desc><desc class="y left">226.2853939606025</desc></g><g class="dots"><circle cx="251.07282565902872" cy="350.67941071053787" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.324318908: 0.4694047691</desc><desc class="x top">251.07282565902872</desc><desc class="y top">350.67941071053787</desc></g><g class="dots"><circle cx="119.06040915828333" cy="486.5547518091239" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2491065327: 0.3697442124</desc><desc class="x top">119.06040915828333</desc><desc class="y top">486.5547518091239</desc></g><g class="dots"><circle cx="73.00314198098302" cy="412.62892946762275" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2228659974: 0.4239666235</desc><desc class="x top">73.00314198098302</desc><desc class="y top">412.62892946762275</desc></g><g class="dots"><circle cx="138.0706159719668" cy="369.0473338349626" r="2.5" class="dot reactive tooltip-trigger"/><desc class="value">0.2599373534: 0.455932439</desc><desc class="x top">138.0706159719668</desc><desc class="y top">369.0473338349626</desc></g></g></g><g transform="translate(84, 46)" class="plot text-overlay"><g class="series serie-0 color-0"/></g><g transform="translate(84, 46)" class="plot tooltip-overlay"><g transform="translate(0 0)" style="opacity: 0" class="tooltip"><rect rx="0" ry="0" width="0" height="0" class="tooltip-box"/><g class="text"/></g></g><g transform="translate(10, 56)" class="legends"><g id="activate-serie-0" class="legend reactive activate-serie"><rect x="0.0" y="1.0" width="12" height="12" class="color-0 reactive"/><text x="17.0" y="11.2">H</text></g></g><g transform="translate(790, 56)" class="legends"/></g></svg>