mirror of
https://github.com/guezoloic/millesima-ai-engine.git
synced 2026-03-28 18:03:47 +00:00
ajout : remplacer appellation par les colonnes indicatrices
This commit is contained in:
15
cleaning.py
15
cleaning.py
@@ -74,3 +74,18 @@ def fill_missing_scores(df: DataFrame) -> DataFrame:
|
||||
|
||||
df_copy = df_copy.drop(columns=temp_cols)
|
||||
return df_copy
|
||||
|
||||
|
||||
def encode_appellation(df: DataFrame, column: str = "Appellation") -> DataFrame:
|
||||
"""
|
||||
Remplace la colonne 'Appellation' par des colonnes indicatrices
|
||||
"""
|
||||
df_copy = df.copy()
|
||||
|
||||
appellations = df_copy[column].astype(str).str.strip()
|
||||
|
||||
appellation_dummies = pd.get_dummies(appellations)
|
||||
|
||||
df_copy = df_copy.drop(columns=[column])
|
||||
|
||||
return df_copy.join(appellation_dummies)
|
||||
|
||||
6
main.py
6
main.py
@@ -59,6 +59,12 @@ def main() -> None:
|
||||
print("\n=== Après remplissage des notes manquantes ===")
|
||||
display_info(df_missing_scores)
|
||||
|
||||
df_ready = encode_appellation(df_missing_scores)
|
||||
save_csv(df_ready, "donnee_ready.csv")
|
||||
print("\n=== Après remplacer la colonne 'Appellation' par des colonnes indicatrices ===")
|
||||
display_info(df_ready)
|
||||
print(df_ready.filter(like="App_").any().head())
|
||||
|
||||
if __name__ == "__main__":
|
||||
try:
|
||||
main()
|
||||
|
||||
Reference in New Issue
Block a user