feat: Add language detection for all text

main.py

@@ -41,3 +41,13 @@ if __name__ == '__main__':
 
     print('Résultat backward ---------------------------------------------------')
     print(f'FR={proba_back_fr}, EN={proba_back_en}, IT={proba_back_it}, Conclusion={max(proba_back_fr, proba_back_en, proba_back_it)}')
+
+    words_text_1 = data_preparation.prepare_data('texte_1.txt')
+    words_text_2 = data_preparation.prepare_data('texte_2.txt')
+    words_text_3 = data_preparation.prepare_data('texte_3.txt')
+
+    text_1_result = utils.forward_detection_with_text(lambda_fr, lambda_en, lambda_it, words_text_1)
+    text_2_result = utils.forward_detection_with_text(lambda_fr, lambda_en, lambda_it, words_text_2)
+    text_3_result = utils.forward_detection_with_text(lambda_fr, lambda_en, lambda_it, words_text_3)
+
+    print(f'texte 1 {text_1_result}, texte 2 {text_2_result}, texte 3 {text_3_result}')

utils/__init__.py

@@ -9,18 +9,18 @@ def normalize_probabilities(prob_fr: float, prob_en: float, prob_it: float, sear
     return searched / sum
 
 
-def forward_detection(hmm_fr: HMM, hmm_en: HMM, hmm_it: HMM, O: str):
+def forward_detection(hmm_fr: HMM, hmm_en: HMM, hmm_it: HMM, O: list[int]) -> str:
     """
     Prints the language detected with forward method
     :param hmm_fr: lambda_fr
     :param hmm_en: lambda_en
     :param hmm_it: lambda_it
     :param O: mot à détecter
-    :return:
+    :return: Le langage détecté
     """
-    res_fr = hmm_fr.forward(O)
+    res_fr, _ = hmm_fr.forward(O)
-    res_en = hmm_en.forward(O)
+    res_en, _ = hmm_en.forward(O)
-    res_it = hmm_it.forward(O)
+    res_it, _ = hmm_it.forward(O)
 
     proba_fr = normalize_probabilities(res_fr, res_en, res_it, res_fr)
     proba_en = normalize_probabilities(res_fr, res_en, res_it, res_en)
@@ -28,21 +28,21 @@ def forward_detection(hmm_fr: HMM, hmm_en: HMM, hmm_it: HMM, O: str):
 
     print(f'FR={proba_fr}, EN={proba_en}, IT={proba_it}')
 
-    probas = [proba_fr, proba_en, proba_it]
+    probas: list[float] = [proba_fr, proba_en, proba_it]
 
     max_prob = proba_fr
     language_index = 0
-    for index in 1..len(probas):
+    for index in range(1, len(probas)):
         if max_prob < probas[index]:
             max_prob = probas[index]
             language_index = index
 
     if language_index == 0:
-        print('Français')
+        return 'Français'
     elif language_index == 1:
-        print('Anglais')
+        return 'Anglais'
     else:
-        print('Italien')
+        return 'Italien'
 
 
 def backward_detection(hmm_fr: HMM, hmm_en: HMM, hmm_it: HMM, O: str):
@@ -68,17 +68,42 @@ def backward_detection(hmm_fr: HMM, hmm_en: HMM, hmm_it: HMM, O: str):
 
     max_prob = proba_fr
     language_index = 0
-    for index in 1..len(probas):
+    for index in range(1,len(probas)):
         if max_prob < probas[index]:
             max_prob = probas[index]
             language_index = index
 
     if language_index == 0:
-        print('Français')
+        return 'Français'
     elif language_index == 1:
-        print('Anglais')
+        return 'Anglais'
     else:
-        print('Italien')
+        return 'Italien'
 
 
+def forward_detection_with_text(hmm_fr: HMM, hmm_en: HMM, hmm_it: HMM, O: list[list[int]]) -> dict:
+    """
 
+    :param hmm_fr: lambda fr
+    :param hmm_en: lambda en
+    :param hmm_it: lambda it
+    :param O: Le texte en version index de l'alphabet
+    :return: Le langage détecté
+    """
+
+    # Count the language détection occurrences. The max count is the answer
+    french_count = 0
+    english_count = 0
+    italian_count = 0
+
+    for word in O:
+        lang = forward_detection(hmm_fr, hmm_en, hmm_it, word)
+        match lang:
+            case 'Français':
+                french_count += 1
+            case 'Anglais':
+                english_count += 1
+            case 'Italien':
+                italian_count += 1
+
+    return {'french': french_count, 'english': english_count, 'italian': italian_count}