connected neural network with dog/cat model

.vscode/settings.json

@@ -0,0 +1,3 @@
+{
+    "git.ignoreLimitWarning": true
+}

classes/minesweeper.py

@@ -3,6 +3,7 @@ import pygame
 from classes import system
 from random import randrange
 from classes import decisionTrees
+from classes.neuralNetwork import NeuralNetwork
 
 pygame.mixer.init()
 
@@ -12,16 +13,18 @@ class NotMine():
     position_y: int
     size: int
     ismine: bool
+    image_path: str
     image: pygame.surface.Surface
     font: pygame.font.Font
     done_text: pygame.surface.Surface
 
-    def __init__(self, position_x, position_y, size, ismine):
+    def __init__(self, position_x, position_y, size):
         self.position_x = position_x
         self.position_y = position_y
         self.size = size
-        self.ismine = ismine
+        self.ismine = False
-        self.image = pygame.image.load("assets/sprites/notmines/" + str(randrange(1, 2)) + ".png")
+        self.image_path = "assets/sprites/notmines/" + str(randrange(1, 3)) + ".jpg"
+        self.image = pygame.image.load(self.image_path)
         self.image = pygame.transform.scale(self.image, (self.size, self.size))
         self.font = pygame.font.SysFont('Comic Sans MS', int(self.size * 0.3))
         self.done_text = self.font.render("", False, (255,0,0))
@@ -37,6 +40,7 @@ class Mine():
     position_y: int
     size: int
     ismine: bool
+    image_path: str
     image: pygame.surface.Surface
     font: pygame.font.Font
     image_text: pygame.surface.Surface
@@ -47,15 +51,16 @@ class Mine():
     weight: float = 1.0
     explosion_timer: int = 100
 
-    def __init__(self, position_x, position_y, size, ismine, difficulty=1, weight=1.0, timer=100):
+    def __init__(self, position_x, position_y, size, difficulty=1, weight=1.0, timer=100):
         self.position_x = position_x
         self.position_y = position_y
         self.size = size
-        self.ismine = ismine
+        self.ismine = True
         self.weight = weight
         self.explosion_timer = timer
         self.difficulty = difficulty
-        self.image = pygame.image.load("assets/sprites/mines/" + str(randrange(1, 2)) + ".png")
+        self.image_path = "assets/sprites/mines/" + str(randrange(1, 3)) + ".jpg"
+        self.image = pygame.image.load(self.image_path)
         self.image = pygame.transform.scale(self.image, (self.size, self.size))
         self.font = pygame.font.SysFont('Comic Sans MS', int(self.size * 0.3))
         self.image_text = self.font.render(str(self.weight), False, (255, 0, 0))
@@ -192,17 +197,15 @@ class Map:
                             break
                     if ok and randrange(10) == 0 and not (i < 2 and j < 3):
                         #zależnie od wylosowanej liczby mina lub niemina
-                        if randrange(0, 10) > 3:
+                        if randrange(0, 10) > 3: #odpowiednia wartość to > 3
-                            ismine = True
                             difficulty = randrange(0, 4) + 1
                             weight = randrange(10, 31) / 10
                             timer = randrange(100, 200)
-                            mine = Mine(j, i, self.tile_size, ismine, difficulty, weight, timer)
+                            mine = Mine(j, i, self.tile_size, difficulty, weight, timer)
                             self.mines.append(mine)
                             self.encounters.append(mine)
                         else:
-                            ismine = False
+                            notmine = NotMine(j, i, self.tile_size)
-                            notmine = NotMine(j, i, self.tile_size, ismine)
                             self.notmines.append(notmine)
                             self.encounters.append(notmine)
 
@@ -304,6 +307,7 @@ class Minesweeper:
         self.image = pygame.transform.scale(self.image, (self.size, self.size))
         self.rotated_image = self.image
         self.rotation_degrees = 0
+        self.neural_network = NeuralNetwork()
 
     def set_map(self, map: Map):
         self.current_map = map
@@ -435,8 +439,8 @@ class Minesweeper:
             if (self.position_x, self.position_y) == (encounter.position_x, encounter.position_y):
                 
                 #tutaj będzie sprawdzanie zdjęcia i na podstawie tego przypisywane true albo false do decisionismine
-                decisionismine = encounter.ismine 
+                decisionismine = self.neural_network.recognize(encounter.image_path)
-
+                
                 #wykryto błędnie
                 if decisionismine != encounter.ismine:
                     print("ERROR: Decision was not right")
@@ -445,6 +449,7 @@ class Minesweeper:
                 #wykryto poprawnie, że mina
                 elif decisionismine:
                     print("Mine? - Yes")
+                    print("")
                     tree = decisionTrees.DecisionTrees()
                     decision = tree.return_predict(model)
                     print("Decision : ", decision, "\n")
@@ -457,6 +462,7 @@ class Minesweeper:
                 #wykryto poprawnie, że niemina
                 else:
                     print("Mine? - No")
+                    print("")
                     encounter.done_text = encounter.font.render("X", False, (255,0,0))
                     self.current_map.encounters.remove(encounter)
                     pygame.mixer.Channel(3).set_volume(0.01)

classes/neuralNetwork.py

@@ -0,0 +1,110 @@
+import pandas as pd
+import tensorflow as tf
+import numpy as np
+import warnings
+import os
+from keras.utils import load_img
+import keras
+from sklearn.model_selection import train_test_split
+from keras.preprocessing.image import ImageDataGenerator
+from keras import Sequential
+from keras.layers import Conv2D, MaxPool2D, Flatten, Dense
+warnings.filterwarnings('ignore')
+
+create_model = False
+learning_sets_path = "data/learning_sets"
+save_model_path = "data/models/true_mine_recognizer.model"
+load_model_path = "data/models/mine_recognizer.model"
+image_size = 128
+
+class NeuralNetwork():
+    def __init__(self):
+        if create_model:
+            input_path = []
+            label = []
+
+            for class_name in os.listdir(learning_sets_path):
+                for path in os.listdir(learning_sets_path+ "/" +class_name):
+                    if class_name == 'mine':
+                        label.append(0)
+                    else:
+                        label.append(1)
+                    input_path.append(os.path.join(learning_sets_path, class_name, path))
+            print(input_path[0], label[0])
+            df = pd.DataFrame()
+            df['images'] = input_path
+            df['label'] = label
+            df = df.sample(frac=1).reset_index(drop=True)
+            df.head()
+            df['label'] = df['label'].astype('str')
+            df.head()
+
+            train, test = train_test_split(df, test_size=0.2, random_state=42)
+
+
+            train_generator = ImageDataGenerator(
+                rescale = 1./255,  
+                rotation_range = 40, 
+                shear_range = 0.2,
+                zoom_range = 0.2,
+                horizontal_flip = True,
+                fill_mode = 'nearest'
+            )
+
+            val_generator = ImageDataGenerator(rescale = 1./255)
+
+            train_iterator = train_generator.flow_from_dataframe(
+                train, 
+                x_col='images', 
+                y_col='label', 
+                target_size=(image_size,image_size), 
+                batch_size=512, 
+                class_mode='binary'
+            )
+
+            val_iterator = val_generator.flow_from_dataframe(
+                test, 
+                x_col='images', 
+                y_col='label', 
+                target_size=(image_size,image_size), 
+                batch_size=512, 
+                class_mode='binary'
+            )
+
+            self.model = Sequential([
+                                Conv2D(16, (3,3), activation='relu', input_shape=(image_size,image_size,3)),
+                                MaxPool2D((2,2)),
+                                Conv2D(32, (3,3), activation='relu'),
+                                MaxPool2D((2,2)),
+                                Conv2D(64, (3,3), activation='relu'),
+                                MaxPool2D((2,2)),
+                                Flatten(),
+                                Dense(512, activation='relu'),
+                                Dense(1, activation='sigmoid')
+            ])
+            self.model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])
+            self.model.summary()
+            self.model.fit(train_iterator, epochs=10, validation_data=val_iterator)
+            self.model.save(save_model_path)
+        else:
+            self.model = keras.models.load_model(load_model_path,
+                                                compile=True
+                                                )
+    def recognize(self, image_path):
+            image = keras.utils.load_img(image_path, target_size=(image_size, image_size))
+            image_array = keras.utils.img_to_array(image)
+            image_array = keras.backend.expand_dims(image_array, 0)
+
+            prediction = self.model.predict(image_array)
+            if prediction[0] > 0.5:
+                predict = "notmine"
+                accuracy = prediction[0] * 100
+            elif prediction[0] <= 0.5:
+                predict = "mine"
+                accuracy = (1 - prediction[0]) * 100
+
+            print("Image: ",image_path," is classified as: ", predict," with: ", accuracy, " accuracy") 
+            if predict == "mine":
+                return True
+            else:
+                return False

outputs/rules/rules.json

@@ -1,31 +1,31 @@
 [
-{"current_level": 1, "leaf_id": "657140df-de8d-11ec-9c63-d43d7ef1576e", "parents": "root", "rule": "else: return 'detonate'", "feature_idx": -1, "feature_name": "", "instances": 200, "metric": 0, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 1, "leaf_id": "3da5aae2-e008-11ec-95c1-d43d7ef1576e", "parents": "root", "rule": "else: return 'detonate'", "feature_idx": -1, "feature_name": "", "instances": 200, "metric": 0, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 1, "leaf_id": "6572790a-de8d-11ec-ac91-d43d7ef1576e", "parents": "root", "rule": "if obj[4]<=80.67436609605278:", "feature_idx": 4, "feature_name": "Random_detonation_chance", "instances": 200, "metric": 0.9964625048848765, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 1, "leaf_id": "3da6bcfb-e008-11ec-ba71-d43d7ef1576e", "parents": "root", "rule": "if obj[4]<=80.67436609605278:", "feature_idx": 4, "feature_name": "Random_detonation_chance", "instances": 200, "metric": 0.9964625048848765, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 2, "leaf_id": "6581b986-de8d-11ec-9198-d43d7ef1576e", "parents": "6572790a-de8d-11ec-ac91-d43d7ef1576e", "rule": "else: return 'detonate'", "feature_idx": -1, "feature_name": "", "instances": 161, "metric": 0, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 2, "leaf_id": "3db49dc5-e008-11ec-b425-d43d7ef1576e", "parents": "3da6bcfb-e008-11ec-ba71-d43d7ef1576e", "rule": "else: return 'detonate'", "feature_idx": -1, "feature_name": "", "instances": 161, "metric": 0, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 2, "leaf_id": "6582f1e8-de8d-11ec-9c53-d43d7ef1576e", "parents": "6572790a-de8d-11ec-ac91-d43d7ef1576e", "rule": "if obj[2]<=0:", "feature_idx": 2, "feature_name": "Protection_from_defuse", "instances": 161, "metric": 0.9203523274205176, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 2, "leaf_id": "3db5af1f-e008-11ec-ba0b-d43d7ef1576e", "parents": "3da6bcfb-e008-11ec-ba71-d43d7ef1576e", "rule": "if obj[2]<=0:", "feature_idx": 2, "feature_name": "Protection_from_defuse", "instances": 161, "metric": 0.9203523274205176, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 3, "leaf_id": "65905ed3-de8d-11ec-b36a-d43d7ef1576e", "parents": "6582f1e8-de8d-11ec-9c53-d43d7ef1576e", "rule": "else: return 'detonate'", "feature_idx": -1, "feature_name": "", "instances": 135, "metric": 0, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 3, "leaf_id": "3dc2a6e1-e008-11ec-ba72-d43d7ef1576e", "parents": "3db5af1f-e008-11ec-ba0b-d43d7ef1576e", "rule": "else: return 'detonate'", "feature_idx": -1, "feature_name": "", "instances": 135, "metric": 0, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 3, "leaf_id": "659196b8-de8d-11ec-b5e3-d43d7ef1576e", "parents": "6582f1e8-de8d-11ec-9c53-d43d7ef1576e", "rule": "if obj[3]<=7:", "feature_idx": 3, "feature_name": "Meters_under_the_ground", "instances": 135, "metric": 0.7364977795505669, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 3, "leaf_id": "3dc39134-e008-11ec-9943-d43d7ef1576e", "parents": "3db5af1f-e008-11ec-ba0b-d43d7ef1576e", "rule": "if obj[3]<=7:", "feature_idx": 3, "feature_name": "Meters_under_the_ground", "instances": 135, "metric": 0.7364977795505669, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 4, "leaf_id": "659b5a72-de8d-11ec-af31-d43d7ef1576e", "parents": "659196b8-de8d-11ec-b5e3-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 116, "metric": 0, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 4, "leaf_id": "3dcd2d59-e008-11ec-9f6f-d43d7ef1576e", "parents": "3dc39134-e008-11ec-9943-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 116, "metric": 0, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 4, "leaf_id": "659c924b-de8d-11ec-ba5c-d43d7ef1576e", "parents": "659196b8-de8d-11ec-b5e3-d43d7ef1576e", "rule": "if obj[1]<=1997.8794790831414:", "feature_idx": 1, "feature_name": "Year(older_more_difficult)", "instances": 116, "metric": 0.3936164041111624, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 4, "leaf_id": "3dce17a5-e008-11ec-acb7-d43d7ef1576e", "parents": "3dc39134-e008-11ec-9943-d43d7ef1576e", "rule": "if obj[1]<=1997.8794790831414:", "feature_idx": 1, "feature_name": "Year(older_more_difficult)", "instances": 116, "metric": 0.3936164041111624, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 5, "leaf_id": "65a396a0-de8d-11ec-a956-d43d7ef1576e", "parents": "659c924b-de8d-11ec-ba5c-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 97, "metric": 0, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 5, "leaf_id": "3dd4f4e4-e008-11ec-9a04-d43d7ef1576e", "parents": "3dce17a5-e008-11ec-acb7-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 97, "metric": 0, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 5, "leaf_id": "65a4a8a9-de8d-11ec-8c86-d43d7ef1576e", "parents": "659c924b-de8d-11ec-ba5c-d43d7ef1576e", "rule": "if obj[5]<=2:", "feature_idx": 5, "feature_name": "Detonation_power_in_m", "instances": 97, "metric": 0.445693177722561, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 5, "leaf_id": "3dd5df33-e008-11ec-a66b-d43d7ef1576e", "parents": "3dce17a5-e008-11ec-acb7-d43d7ef1576e", "rule": "if obj[5]<=2:", "feature_idx": 5, "feature_name": "Detonation_power_in_m", "instances": 97, "metric": 0.445693177722561, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 6, "leaf_id": "65a851b0-de8d-11ec-ad73-d43d7ef1576e", "parents": "65a4a8a9-de8d-11ec-8c86-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 65, "metric": 0, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 6, "leaf_id": "3dd9615a-e008-11ec-a021-d43d7ef1576e", "parents": "3dd5df33-e008-11ec-a66b-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 65, "metric": 0, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 6, "leaf_id": "65a9899b-de8d-11ec-af58-d43d7ef1576e", "parents": "65a4a8a9-de8d-11ec-8c86-d43d7ef1576e", "rule": "if obj[0]>3:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0.33352200393097664, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 6, "leaf_id": "3dda7234-e008-11ec-bf9b-d43d7ef1576e", "parents": "3dd5df33-e008-11ec-a66b-d43d7ef1576e", "rule": "if obj[0]>3:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0.33352200393097664, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 7, "leaf_id": "65a9899c-de8d-11ec-a96c-d43d7ef1576e", "parents": "65a9899b-de8d-11ec-af58-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0, "return_statement": 1, "tree_id": 0}, 
+{"current_level": 7, "leaf_id": "3dda7235-e008-11ec-bf80-d43d7ef1576e", "parents": "3dda7234-e008-11ec-bf9b-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0, "return_statement": 1, "tree_id": 0}, 
-{"current_level": 6, "leaf_id": "65abfae1-de8d-11ec-aaaa-d43d7ef1576e", "parents": "65a4a8a9-de8d-11ec-8c86-d43d7ef1576e", "rule": "if obj[0]<=3:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0.33352200393097664, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 6, "leaf_id": "3de6589c-e008-11ec-b9e2-d43d7ef1576e", "parents": "3dd5df33-e008-11ec-a66b-d43d7ef1576e", "rule": "if obj[0]<=3:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0.33352200393097664, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 7, "leaf_id": "65abfae2-de8d-11ec-8f35-d43d7ef1576e", "parents": "65abfae1-de8d-11ec-aaaa-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0, "return_statement": 1, "tree_id": 0}, 
+{"current_level": 7, "leaf_id": "3de6589d-e008-11ec-ac82-d43d7ef1576e", "parents": "3de6589c-e008-11ec-b9e2-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 65, "metric": 0, "return_statement": 1, "tree_id": 0}, 
-{"current_level": 5, "leaf_id": "65b06753-de8d-11ec-9af2-d43d7ef1576e", "parents": "659c924b-de8d-11ec-ba5c-d43d7ef1576e", "rule": "if obj[5]>2:", "feature_idx": 5, "feature_name": "Detonation_power_in_m", "instances": 97, "metric": 0.445693177722561, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 5, "leaf_id": "3dfdfd61-e008-11ec-a211-d43d7ef1576e", "parents": "3dce17a5-e008-11ec-acb7-d43d7ef1576e", "rule": "if obj[5]>2:", "feature_idx": 5, "feature_name": "Detonation_power_in_m", "instances": 97, "metric": 0.445693177722561, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 6, "leaf_id": "65b41086-de8d-11ec-b9c3-d43d7ef1576e", "parents": "65b06753-de8d-11ec-9af2-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 32, "metric": 0, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 6, "leaf_id": "3e01a611-e008-11ec-910f-d43d7ef1576e", "parents": "3dfdfd61-e008-11ec-a211-d43d7ef1576e", "rule": "else: return 'defuse'", "feature_idx": -1, "feature_name": "", "instances": 32, "metric": 0, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 6, "leaf_id": "65b548ec-de8d-11ec-8daf-d43d7ef1576e", "parents": "65b06753-de8d-11ec-9af2-d43d7ef1576e", "rule": "if obj[0]<=7:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0.6252624052234231, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 6, "leaf_id": "3e029055-e008-11ec-94d4-d43d7ef1576e", "parents": "3dfdfd61-e008-11ec-a211-d43d7ef1576e", "rule": "if obj[0]<=7:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0.6252624052234231, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 7, "leaf_id": "65b548ed-de8d-11ec-a6f9-d43d7ef1576e", "parents": "65b548ec-de8d-11ec-8daf-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0, "return_statement": 1, "tree_id": 0}, 
+{"current_level": 7, "leaf_id": "3e029056-e008-11ec-80d0-d43d7ef1576e", "parents": "3e029055-e008-11ec-94d4-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0, "return_statement": 1, "tree_id": 0}, 
-{"current_level": 6, "leaf_id": "65b792b4-de8d-11ec-ae1b-d43d7ef1576e", "parents": "65b06753-de8d-11ec-9af2-d43d7ef1576e", "rule": "if obj[0]>7:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0.6252624052234231, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 6, "leaf_id": "3e0e9dd1-e008-11ec-8ed3-d43d7ef1576e", "parents": "3dfdfd61-e008-11ec-a211-d43d7ef1576e", "rule": "if obj[0]>7:", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0.6252624052234231, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 7, "leaf_id": "65b792b5-de8d-11ec-8bb4-d43d7ef1576e", "parents": "65b792b4-de8d-11ec-ae1b-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0, "return_statement": 1, "tree_id": 0}, 
+{"current_level": 7, "leaf_id": "3e0e9dd2-e008-11ec-9bd3-d43d7ef1576e", "parents": "3e0e9dd1-e008-11ec-8ed3-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 0, "feature_name": "Size(bigger_more_difficult)", "instances": 32, "metric": 0, "return_statement": 1, "tree_id": 0}, 
-{"current_level": 4, "leaf_id": "65be48e0-de8d-11ec-b4ae-d43d7ef1576e", "parents": "659196b8-de8d-11ec-b5e3-d43d7ef1576e", "rule": "if obj[1]>1997.8794790831414:", "feature_idx": 1, "feature_name": "Year(older_more_difficult)", "instances": 116, "metric": 0.3936164041111624, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 4, "leaf_id": "3e324f8e-e008-11ec-a2cf-d43d7ef1576e", "parents": "3dc39134-e008-11ec-9943-d43d7ef1576e", "rule": "if obj[1]>1997.8794790831414:", "feature_idx": 1, "feature_name": "Year(older_more_difficult)", "instances": 116, "metric": 0.3936164041111624, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 5, "leaf_id": "65be48e1-de8d-11ec-9136-d43d7ef1576e", "parents": "65be48e0-de8d-11ec-b4ae-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 1, "feature_name": "Year(older_more_difficult)", "instances": 116, "metric": 0, "return_statement": 1, "tree_id": 0}, 
+{"current_level": 5, "leaf_id": "3e324f8f-e008-11ec-b0e5-d43d7ef1576e", "parents": "3e324f8e-e008-11ec-a2cf-d43d7ef1576e", "rule": "return 'defuse'", "feature_idx": 1, "feature_name": "Year(older_more_difficult)", "instances": 116, "metric": 0, "return_statement": 1, "tree_id": 0}, 
-{"current_level": 3, "leaf_id": "65c2b4d6-de8d-11ec-89c3-d43d7ef1576e", "parents": "6582f1e8-de8d-11ec-9c53-d43d7ef1576e", "rule": "if obj[3]>7:", "feature_idx": 3, "feature_name": "Meters_under_the_ground", "instances": 135, "metric": 0.7364977795505669, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 3, "leaf_id": "3e4a6970-e008-11ec-82ec-d43d7ef1576e", "parents": "3db5af1f-e008-11ec-ba0b-d43d7ef1576e", "rule": "if obj[3]>7:", "feature_idx": 3, "feature_name": "Meters_under_the_ground", "instances": 135, "metric": 0.7364977795505669, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 4, "leaf_id": "65c2b4d7-de8d-11ec-8575-d43d7ef1576e", "parents": "65c2b4d6-de8d-11ec-89c3-d43d7ef1576e", "rule": "return 'detonate'", "feature_idx": 3, "feature_name": "Meters_under_the_ground", "instances": 135, "metric": 0, "return_statement": 1, "tree_id": 0}, 
+{"current_level": 4, "leaf_id": "3e4a6971-e008-11ec-b6a1-d43d7ef1576e", "parents": "3e4a6970-e008-11ec-82ec-d43d7ef1576e", "rule": "return 'detonate'", "feature_idx": 3, "feature_name": "Meters_under_the_ground", "instances": 135, "metric": 0, "return_statement": 1, "tree_id": 0}, 
-{"current_level": 2, "leaf_id": "65c72151-de8d-11ec-adc4-d43d7ef1576e", "parents": "6572790a-de8d-11ec-ac91-d43d7ef1576e", "rule": "if obj[2]>0:", "feature_idx": 2, "feature_name": "Protection_from_defuse", "instances": 161, "metric": 0.9203523274205176, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 2, "leaf_id": "3e625c50-e008-11ec-8286-d43d7ef1576e", "parents": "3da6bcfb-e008-11ec-ba71-d43d7ef1576e", "rule": "if obj[2]>0:", "feature_idx": 2, "feature_name": "Protection_from_defuse", "instances": 161, "metric": 0.9203523274205176, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 3, "leaf_id": "65c72152-de8d-11ec-8a44-d43d7ef1576e", "parents": "65c72151-de8d-11ec-adc4-d43d7ef1576e", "rule": "return 'detonate'", "feature_idx": 2, "feature_name": "Protection_from_defuse", "instances": 161, "metric": 0, "return_statement": 1, "tree_id": 0}, 
+{"current_level": 3, "leaf_id": "3e625c51-e008-11ec-bcac-d43d7ef1576e", "parents": "3e625c50-e008-11ec-8286-d43d7ef1576e", "rule": "return 'detonate'", "feature_idx": 2, "feature_name": "Protection_from_defuse", "instances": 161, "metric": 0, "return_statement": 1, "tree_id": 0}, 
-{"current_level": 1, "leaf_id": "65cb8dc2-de8d-11ec-b823-d43d7ef1576e", "parents": "root", "rule": "if obj[4]>80.67436609605278:", "feature_idx": 4, "feature_name": "Random_detonation_chance", "instances": 200, "metric": 0.9964625048848765, "return_statement": 0, "tree_id": 0}, 
+{"current_level": 1, "leaf_id": "3e7da9c6-e008-11ec-9254-d43d7ef1576e", "parents": "root", "rule": "if obj[4]>80.67436609605278:", "feature_idx": 4, "feature_name": "Random_detonation_chance", "instances": 200, "metric": 0.9964625048848765, "return_statement": 0, "tree_id": 0}, 
-{"current_level": 2, "leaf_id": "65cb8dc3-de8d-11ec-a99c-d43d7ef1576e", "parents": "65cb8dc2-de8d-11ec-b823-d43d7ef1576e", "rule": "return 'detonate'", "feature_idx": 4, "feature_name": "Random_detonation_chance", "instances": 200, "metric": 0, "return_statement": 1, "tree_id": 0}
+{"current_level": 2, "leaf_id": "3e7da9c7-e008-11ec-81ca-d43d7ef1576e", "parents": "3e7da9c6-e008-11ec-9254-d43d7ef1576e", "rule": "return 'detonate'", "feature_idx": 4, "feature_name": "Random_detonation_chance", "instances": 200, "metric": 0, "return_statement": 1, "tree_id": 0}
 ]

requirements.txt

@@ -1,2 +1,4 @@
 pygame
-chefboost
+chefboost
+tensorflow #--upgrade
+sklearn