decision tree integration + some fixes

requirements.txt

@@ -5,4 +5,6 @@ scikit-learn
 IPython
 pydotplus
 torch
-tensorflow
+tensorflow
+Pillow
+keras

src/ANN/test-predict.py

@@ -13,9 +13,9 @@ model = load_model(model_path)
 class_names = os.listdir(dataset_dir)
 
 # Путь к файло проверки
-dataset_dir = "rdy-dataset\\test\\rose\\377277099_544769262c_c.jpg"
+dataset_dir = "dataset\iris\\71766650_e2d423bea7_c.jpg"
 
-image_path = 'rdy-dataset\\test\\rose\\377277099_544769262c_c.jpg'  # Replace with the path to your image
+image_path = 'dataset\iris\\71766650_e2d423bea7_c.jpg'  # Replace with the path to your image
 image = cv2.imread(image_path)
 image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)  # Convert image to RGB format
 image = cv2.resize(image, (64, 64))  # Resize image to match the input size of the model

src/Field.py

@@ -60,6 +60,10 @@ class Field:
             for y in range(len(self.Nodes[0])):
                 print(self.Nodes[x][y].Values, end=' ')
 
+    def set_type(self, pos, type):
+        X,Y = pos
+        self.Nodes[X][Y].Type = type
+
 
 def Neighbors(field, pos):
     x, y, dir = pos

src/main.py

@@ -1,5 +1,9 @@
+import pandas as pd
 import pygame
 import random
+
+from sklearn.preprocessing import LabelEncoder
+
 #import Flower
 import Beehive
 import Frames
@@ -240,9 +244,26 @@ print(len(check_path_A))
 current_node = 0
 step = 1
 comeback = False
-
-
-
+weather = random.choice(['Sunny', 'Rainy', 'Cloudy'])
+temperature = random.choice(['High', 'Medium', 'Low'])
+windspeed = random.choice(['Strong', 'Weak'])
+pollenavailability = random.choice(['Low', 'Medium', 'High'])
+beepopulation = random.choice(['Low','Moderate', 'High'])
+timeofday = random.choice(['Morning', 'Afternoon', 'Evening'])
+if len(check_path_A)>80:
+    flowerdistance = 'Far'
+elif len(check_path_A)>50:
+    flowerdistance = 'Medium'
+else:
+    flowerdistance = 'Close'
+humidity = random.choice(['High','Normal'])
+predictiondata = [[weather, temperature, windspeed, pollenavailability, beepopulation, timeofday, flowerdistance, humidity]]
+feature_cols = ['Weather','Temperature','WindSpeed','PollenAvailability','BeePopulation','TimeOfDay','FlowerDistance','Humidity']
+le = LabelEncoder()
+le.fit(feature_cols)
+encoded_data = pd.DataFrame(predictiondata, columns=feature_cols).apply(le.fit_transform)
+prediction = clf.predict(encoded_data)
+font = pygame.font.Font(None,32)
 while True:
     # check for events
     for event in pygame.event.get():
@@ -268,6 +289,7 @@ while True:
 
     dis.blit(bee_image, (bee_x * tileSize, bee_y * tileSize))
     pygame.display.update()
+
     if (bee_x,bee_y) in flower_positions:
         if field.get_type((bee_x,bee_y)) == FLOWER:
             if chek_flower(pygame.image.load(field.get_image((bee_x,bee_y)))) not in ['western-poison-oak', 'water_lily']:
@@ -276,7 +298,12 @@ while True:
                 field.set_type((bee_x,bee_y), SHORT)
             else:
                 field.set_type((bee_x,bee_y), SHORT)
-
+    if prediction[0] == 0:
+        text = font.render('The bee refused to fly out for honey', True, (255, 0, 0))
+        text_rect = text.get_rect(center=(disX // 2, disY // 2))
+        dis.blit(text, text_rect)
+        pygame.display.update()
+        pygame.time.wait(10000000)
 
     clock.tick(10)