l'appli console marche, bonne nuit aurel

calculsIA.py

@@ -2,41 +2,64 @@ import pandas as pd
 import numpy as np
 import matplotlib.pyplot as plt
 import sklearn as sk
+from sklearn.model_selection import train_test_split
+from sklearn.linear_model import LinearRegression
+from sklearn.metrics import mean_squared_error, mean_absolute_error
 
-#dfRatingsTropGrand = pd.read_csv("processedData/actorsRatingsPerMovie.tsv",sep='\t')
+# dfRatingsTropGrand = pd.read_csv("processedData/actorsRatingsPerMovie.tsv",sep='\t')
-#tconst	ratings	actorNames	averageRatingMovie
+# tconst	ratings	actorNames	averageRatingMovie
 
-#dfRatings = dfRatingsTropGrand[dfRatingsTropGrand['ratings'].apply(lambda x: len(eval(x)) >= 4)]
+# dfRatings = dfRatingsTropGrand[dfRatingsTropGrand['ratings'].apply(lambda x: len(eval(x)) >= 4)]
 
-#dfRatings.to_csv("processedData/actorsRatingsPerMovieGoodToUse.tsv", index=False, sep="\t")
+# dfRatings.to_csv("processedData/actorsRatingsPerMovieGoodToUse.tsv", index=False, sep="\t")
 dfRatings = pd.read_csv("processedData/actorsRatingsPerMovieGoodToUse.tsv", sep="\t")
-
+dfActeurs = pd.read_csv("processedData/actorsRatingsGroupedWithName.tsv", sep="\t")
-#listMovies = dfRatings.head(1000)['tconst'].values
+print("Veuillez entrer un entier positif inférieur ou égal à ",len(dfRatings))
-listMovies = dfRatings['tconst'].values
+print("(Plus le nombre est petit, le temps de préparation sera moins long, mais la précision du modèle sera plus petite)")
+val = input(": ")
+val = int(val)
+listMovies = dfRatings.sample(val)['tconst'].values
+# listMovies = dfRatings['tconst'].values
 
 listRatingsA = []
 listRatingsM = []
 datas = []
 nbDiese = 0
 for i in range(len(listMovies)):
-	print(i/len(listMovies)*100,"%", end="\r")
+    valPrct = i / len(listMovies) * 100
+    print("{:.2f}".format(valPrct), "%", end="\r")
     film = listMovies[i]
-	bob = (dfRatings.averageRatingMovie.loc[dfRatings.tconst == film].values[0],eval(dfRatings.ratings.loc[dfRatings.tconst == film].values[0]))
+    bob = (dfRatings.averageRatingMovie.loc[dfRatings.tconst == film].values[0],
+           eval(dfRatings.ratings.loc[dfRatings.tconst == film].values[0]))
     listRatingsA.append(bob[1][:4])
     listRatingsM.append(bob[0])
 print("")
 
-from sklearn.model_selection import train_test_split
+x = listRatingsA
+y = listRatingsM
 
-x=listRatingsA
+xtrain, xtest, ytrain, ytest = train_test_split(x, y, test_size=0.3)
-y=listRatingsM
 
-xtrain, xtest, ytrain, ytest = train_test_split(x,y,test_size=0.3)
+lnrg = LinearRegression()
 
+# clf = lnrg.fit(xtrain,ytrain)
 xtrain = np.array(xtrain)
-from sklearn.linear_model import LinearRegression
+clf = lnrg.fit(x, y)
+predictions = lnrg.predict(xtest)
+print("\nPréparation du modèle de regréssion linéaire terminée\n")
 
-lnrg = LinearRegression()
+print('Erreur quadratique : ', mean_squared_error(ytest, predictions))
+print('Écart moyen : ', mean_absolute_error(ytest, predictions),"\n")
 
-#clf = lnrg.fit(xtrain,ytrain)
+def calculPrevision(listNomsActeurs):
-clf = lnrg.fit(x,y)
+    if len(listNomsActeurs) == 4:
+        print('\nPrédiction en cours...\n')
+        notesActeurs = []
+        for nom in listNomsActeurs:
+            note = dfActeurs.loc[dfActeurs.primaryName == nom].averageRatingMean.values[0]
+            print(nom, " a pour note moyenne : ", note)
+            notesActeurs.append(note)
+        prediction = clf.predict([notesActeurs])[0]
+        print("\nNote prédite : ", "{:.2f}".format(prediction), "\n")
+    else:
+        print("La liste d'acteurs n'est pas de la bonne taille")

interface.py

@@ -3,6 +3,7 @@ import os
 from typing import List
 import numpy as np
 import pandas as pd
+from calculsIA import *
 
 def levenshtein_distance(s1, s2):
     if len(s1) < len(s2):
@@ -23,6 +24,7 @@ def levenshtein_distance(s1, s2):
 
     return previous_row[-1]
 
+
 def find_closest_actor_name(input_name, actor_names):
     closest_name = None
     min_distance = float('inf')
@@ -34,12 +36,14 @@ def find_closest_actor_name(input_name, actor_names):
             closest_name = actor_name
     return (closest_name)
 
+
 def getUniqueActorNames(filePath):
     # Lire le fichier TSV
     df = pd.read_csv(filePath, sep='\t')
     actor_names = df['primaryName'].tolist()
     return (actor_names)
 
+
 def saveUniqueActorsSorted(inputFilePath, outputFilePath):
     # Lire le fichier TSV
     df = pd.read_csv(inputFilePath, sep='\t')
@@ -63,6 +67,7 @@ def ask_user_verification(actor_name):
     response = input(f"Est-ce que vous vouliez dire {actor_name}? (Oui/Non) ")
     return response.lower() in ['oui', 'o', 'yes', 'y']
 
+
 def add_actor(actor_names):
     user_input = input("Entrez le nom de l'acteur à ajouter : ")
     closest_name = find_closest_actor_name(user_input, actor_names)
@@ -75,6 +80,7 @@ def add_actor(actor_names):
         else:
             return None
 
+
 if __name__ == "__main__":
     mustContinue = True
     if not os.path.exists("processedData/uniqueActorNames.tsv"):
@@ -95,8 +101,8 @@ if __name__ == "__main__":
         if choice == '1':
             result = add_actor(actor_names)
             if result:
-                selectedActorNames.append(result[0])
+                selectedActorNames.append(result)
-                print(f"{result[0]} a été ajouté à la liste.")
+                print(f"{result} a été ajouté à la liste.")
         elif choice == '2':
             print("Liste des acteurs sélectionnés :")
             for actor in selectedActorNames:
@@ -105,8 +111,7 @@ if __name__ == "__main__":
             selectedActorNames.clear()
             print("La liste a été vidée.")
         elif choice == '4':
-            # TODO: Implémenter la comparaison
+            calculPrevision(selectedActorNames)
-            print("Lancement de la prévision... (TODO)")
         elif choice == '5':
             print("Au revoir !")
             mustContinue = False