Mise à jour des données du chapitre Pandas

.gitignore

@@ -56,3 +56,5 @@ jsconfig.json
 /.luarc.json
 
 **/*.quarto_ipynb
+*.quarto_ipynb*
+.vscode/

_quarto.yml

@@ -5,10 +5,10 @@ project:
     - 404.qmd
     - content/getting-started/index.qmd
     - content/manipulation/index.qmd
+    - content/manipulation/02_pandas_suite.qmd
     - content/visualisation/index.qmd
     - content/getting-started/01_environment.qmd
     - content/modelisation/index.qmd
-    - content/modelisation/4_featureselection.qmd
     - content/NLP/index.qmd
     - content/annexes/corrections.qmd
     - content/annexes/evaluation.qmd

content/manipulation/02_pandas_suite/_exo3.qmd

@@ -16,7 +16,7 @@ On se focalise temporairement sur les observations où le libellé comporte plus
 
 * _Question 4_. Pour mieux y voir, réordonner la base obtenue par order alphabétique.
 
-* _Question 5_. Déterminer la taille moyenne (variable nombre de personnes: `NBPERSMENFISC16`) et quelques statistiques descriptives de ces données.
+* _Question 5_. Déterminer la taille moyenne (variable nombre de personnes: `POPULATION`) et quelques statistiques descriptives de ces données.
 Comparer aux mêmes statistiques sur les données où libellés et codes communes coïncident.
 
 * _Question 6_. Vérifier les grandes villes (plus de 100 000 personnes),
@@ -46,7 +46,7 @@ We temporarily focus on observations where the label involves more than two diff
 
 * _Question 4_. To get a better view, reorder the obtained dataset alphabetically.
 
-* _Question 5_. Determine the average size (variable number of people: `NBPERSMENFISC16`) and some descriptive statistics of this data. Compare it to the same statistics on the data where labels and commune codes coincide.
+* _Question 5_. Determine the average size (variable number of people: `POPULATION`) and some descriptive statistics of this data. Compare it to the same statistics on the data where labels and commune codes coincide.
 
 * _Question 6_. Check the major cities (more than 100,000 people) for the proportion of cities where the same name is associated with different commune codes.
 

content/manipulation/02_pandas_suite/_exo3_solution.qmd

@@ -39,30 +39,30 @@ filosofi.loc[
 print(10*"--" + "Communes dupliquées" + 10*"--")
 print(
   filosofi.loc[
-    filosofi['LIBGEO'].isin(doublons['LIBGEO']), 'NBPERSMENFISC16'
+    filosofi['LIBGEO'].isin(doublons['LIBGEO']), 'POPULATION'
     ].describe()
   )
 print(10*"--" + "Communes non dupliquées" + 10*"--")
 print(
   filosofi.loc[
-    ~filosofi['LIBGEO'].isin(doublons['LIBGEO']), 'NBPERSMENFISC16'
+    ~filosofi['LIBGEO'].isin(doublons['LIBGEO']), 'POPULATION'
     ].describe()
   )
 ```
 
 ```{python}
 #| output: false
 # Question 6
-emissions_big_city = filosofi.loc[filosofi['NBPERSMENFISC16']>100000].copy()
+emissions_big_city = filosofi.loc[filosofi['POPULATION']>100000].copy()
 emissions_big_city['probleme'] = emissions_big_city['LIBGEO'].isin(doublons['LIBGEO'])
-emissions_big_city['probleme'].mean() 
+emissions_big_city['probleme'].mean()
 emissions_big_city[emissions_big_city['probleme']]
-print(100*emissions_big_city['probleme'].mean()) #8,33 %
+print(100*emissions_big_city['probleme'].mean()) #9,52 %
 ```
 
 ```{python}
 #| output: false
 # Question 7
 filosofi[filosofi['LIBGEO'] == 'Montreuil']
-filosofi[filosofi['LIBGEO'].str.contains('Saint-Denis')].head(10)
+filosofi[filosofi['LIBGEO'].str.contains('Saint-Denis', na=False)].head(10)
 ```

content/manipulation/02_pandas_suite/_exo4_solution.qmd

@@ -2,15 +2,16 @@
 #| output: false
 # Question 1
 
-emissions['emissions'] = emissions.sum(axis = 1, numeric_only = True)
+emissions_totales_commune = emissions.sum(axis = 1, numeric_only = True)
 ```
 
 ```{python}
 #| output: false
 # Question 2
 
 emissions_merged = (
-    emissions.reset_index()
+    emissions.assign(emissions = emissions_totales_commune)
+    .reset_index()
     .merge(filosofi, left_on = "INSEE commune", right_on = "CODGEO")
 )
 ```
@@ -19,7 +20,8 @@ emissions_merged = (
 #| output: false
 # Question 3
 
-emissions_merged['empreinte'] = emissions_merged['emissions']/emissions_merged['NBPERSMENFISC16']
+emissions_merged = emissions_merged.loc[emissions_merged['POPULATION'] > 0]
+emissions_merged['empreinte'] = emissions_merged['emissions']/emissions_merged['POPULATION']
 emissions_merged['empreinte'] = emissions_merged['empreinte'].astype(float)
 ```
 

content/manipulation/02_pandas_suite/_exo5_preliminary.qmd

@@ -10,21 +10,23 @@ To ensure you are able to complete the next exercise, here is the dataframe requ
 
 ```{python}
 #| echo: true
-emissions['emissions'] = emissions.sum(axis = 1, numeric_only = True)
+emissions_totales_commune = emissions.sum(axis = 1, numeric_only = True)
 
 emissions_merged = (
-    emissions.reset_index()
+    emissions.assign(emissions = emissions_totales_commune)
+    .reset_index()
     .merge(filosofi, left_on = "INSEE commune", right_on = "CODGEO")
 )
-emissions_merged['empreinte'] = emissions_merged['emissions']/emissions_merged['NBPERSMENFISC16']
+emissions_merged = emissions_merged.loc[emissions_merged['POPULATION'] > 0]
+emissions_merged['empreinte'] = emissions_merged['emissions']/emissions_merged['POPULATION']
 emissions_merged['empreinte'] = emissions_merged['empreinte'].astype(float)
-``` 
+```
 
 ```{python}
 #| echo: true
 emissions_table = (
     emissions_merged
-    .rename(columns={"dep_y": "dep", "NBPERSMENFISC16": "population", "MED16": "revenu"})
+    .rename(columns={"dep_y": "dep", "POPULATION": "population", "NIVVIE_MEDIAN": "revenu"})
     .groupby("dep")
     .agg({"empreinte": "sum", "revenu": "median", "population": "sum"}) #pas vraiment le revenu médian
     .reset_index()