decision tree update

Co-authored-by: Sebastian Piotrowski <sebpio@st.amu.edu.pl>
Co-authored-by: Marcin Matoga <marmat35@st.amu.edu.pl>
Co-authored-by: Ladislaus3III <Ladislaus3III@users.noreply.github.com>
This commit is contained in:
eugenep 2021-05-15 21:04:51 +02:00
parent 74788bbc94
commit f0e0d4f4c1
8 changed files with 365 additions and 1042 deletions

4
.gitignore vendored
View File

@ -1,2 +1,6 @@
# ignore pycache
__pycache__/
# ignore pdf files
*.pdf
data

1200
dane.txt

File diff suppressed because it is too large Load Diff

BIN
images/robot3.bmp Normal file

Binary file not shown.

After

Width:  |  Height:  |  Size: 1.9 MiB

46
learning.py Normal file
View File

@ -0,0 +1,46 @@
from sklearn import tree
import graphviz
class Learning():
X = []
Y = []
clf = tree.DecisionTreeClassifier()
def load_data(self):
file = open('dane.txt', "r")
data_str = []
data_str_separated = []
data = []
for lines in file:
s = file.readline()
data_str.append(s.replace(' \n', ''))
for lines in data_str:
s = lines.split(' ')
data_str_separated.append(s)
for lines in data_str_separated:
line = list(map(int, lines))
data.append(line)
for lines in data:
self.Y.append(lines[8])
lines.pop()
self.X.append(lines)
def learn(self):
#clf = tree.DecisionTreeClassifier()
self.clf = self.clf.fit(self.X, self.Y)
def draw_tree(self):
dot_data = tree.export_graphviz(self.clf, out_file=None, filled=True, class_names= ['1', '2', '3', '4', '5'], rounded=True, special_characters=True)
graph = graphviz.Source(dot_data)
graph.render("data")
def predict(self, param_array):
print(self.clf.predict([param_array]))
if self.clf.predict([param_array]) == 1:
print("oflagowac")
if self.clf.predict([param_array]) == 2:
print("zdetonowac")
if self.clf.predict([param_array]) == 3:
print("sprzedac na allegro")
if self.clf.predict([param_array]) == 4:
print("sprzedac na czarnym rynku")
if self.clf.predict([param_array]) == 5:
print("obejrzec")

33
main.py
View File

@ -14,7 +14,8 @@ from settings import *
from sprites import *
from graphsearch import *
from astar2 import *
from generate_examples import *
from pizza import *
from learning import *
class Game:
@ -26,16 +27,18 @@ class Game:
pg.key.set_repeat(500, 100)
self.load_data()
self.wentyl_bezpieczenstwa = 0
#ExampleGenerator.generate(ExampleGenerator)
def load_data(self):
game_folder = path.dirname(__file__)
self.map_data = []
with open(path.join(game_folder, 'map.txt'), 'rt') as f:
with open(path.join(game_folder, 'main_map.txt'), 'rt') as f:
for line in f:
self.map_data.append(line)
def new(self):
# initialize all variables and do all the setup for a new game
#glebokosc rozmiar masa moc szkodliwosc zabawka stan teledysk
self.all_sprites = pg.sprite.Group()
self.walls = pg.sprite.Group()
self.mines = pg.sprite.Group()
@ -44,6 +47,7 @@ class Game:
for col, tile in enumerate(tiles):
if tile == '2':
Mine(self, col, row)
Mine.set_parameters(Mine,-5,32,6,7,1,0,0,0)
if tile == '3':
Bomb(self, col, row)
if tile == '4':
@ -111,21 +115,27 @@ class Game:
self.player.parse_maze_moves()
self.i_like_to_move_it()
self.wentyl_bezpieczenstwa = 1
if event.key == pg.K_F3 and self.wentyl_bezpieczenstwa == 0:
if event.key == pg.K_F3:
pg.event.clear()
player_moves = BFS.run()
self.graph_move(player_moves)
self.wentyl_bezpieczenstwa = 1
#self.wentyl_bezpieczenstwa = 1
if event.key == pg.K_F4 and self.wentyl_bezpieczenstwa == 0:
pg.event.clear()
#print("test1")
agent = SweeperAgent()
player_moves = SweeperAgent.run(agent)
self.graph_move(player_moves)
self.wentyl_bezpieczenstwa = 1
# Test.run()
if event.key == pg.K_F5:
ExampleGenerator.generate()
print("lol xD")
pg.event.clear()
if event.key == pg.K_F6:
X = [[6,15,31,5,0,0,1,0],[6,36,5,8,1,1,1,0],
pg.event.clear()
#self.wentyl_bezpieczenstwa = 1
"""X = [[6,15,31,5,0,0,1,0],[6,36,5,8,1,1,1,0],
[13,40,29,3,0,0,0,0], [15,25,23,7,0,0,1,1],
[17,34,5,5,1,1,0,1], [3,9,28,8,0,0,1,0],
[18,38,43,6,1,1,1,1], [3,8,53,4,1,1,1,1],
@ -133,7 +143,16 @@ class Game:
Y = [1,5,1,1,4,1,2,2,1,1]
clf = tree.DecisionTreeClassifier()
clf = clf.fit(X,Y)
tree.plot_tree(clf)
tree.plot_tree(clf)"""
"""my_learning = Learning()
my_learning.load_data()
my_learning.learn()
my_learning.draw_tree()
my_learning.predict()"""

7
main_map.txt Normal file
View File

@ -0,0 +1,7 @@
.>.p...
###.ppp
....p..
...pp.p
..p###.
..p#2#.
..ppp..

View File

@ -4,11 +4,12 @@ import random
class ExampleGenerator():
@staticmethod
def generate():
list = [[]]
for i in range(1000):
for i in range(200):
while True:
c = [random.randrange(0, 21), random.randrange(5, 41), random.randrange(1, 61), random.randrange(0, 11), random.randrange(0, 2), random.randrange(0, 2), random.randrange(0, 2), random.randrange(0, 2)]
if c in list:
@ -18,42 +19,43 @@ class ExampleGenerator():
list.append(c)
f = open("dane.txt", "w")
#f1 = open("dane_slownie.txt", "w")
#glebokosc rozmiar masa moc szkodliwosc zabawka stan teledysk
for i in range(len(list)):
if i != 0:
#glebokosc
if list[i][0] <= 19:
#rozmiar
if list[i][1] <= 39:
#zabawka
if list[i][5] == 1:
#teledysk
if list[i][7] == 1:
list[i].append(5)
else:
list[i].append(3)
else:
#glebokosc
if list[i][0] >= 19:
list[i].append(1)
else:
#masa
if list[i][2] <= 58:
#moc
if list[i][3] <= 9:
#szkodliwosc
if list[i][4] == 1:
#zabawka
if list[i][5] == 1:
if list[i][2] >= 58:
list[i].append(1)
else:
#rozmiar
if list[i][1] >= 39:
list[i].append(1)
else:
#moc
if list[i][3] >= 9:
list[i].append(1)
else:
#szkodliwosc
if list[i][4] == 1:
list[i].append(1)
else:
#stan
if list[i][6] == 1:
#teledysk
if list[i][7] == 1:
list[i].append(2)
else:
list[i].append(5)
else:
list[i].append(4)
else:
list[i].append(3)
else:
list[i].append(1)
else:
list[i].append(1)
else:
list[i].append(1)
else:
list[i].append(1)
else:
list[i].append(1)
else:
list[i].append(2)
for i in range(len(list)):
if i == 0:

View File

@ -4,6 +4,7 @@ import ctypes
import ast
from settings import *
from maze import *
from learning import *
class Player(pg.sprite.Sprite):
def __init__(self, game, x, y, direction = 'Right'):
@ -11,8 +12,8 @@ class Player(pg.sprite.Sprite):
pg.sprite.Sprite.__init__(self, self.groups)
self.game = game
#self.image = pg.Surface((TILESIZE, TILESIZE))
self.image = pg.image.load('images/robot2.bmp')
self.baseImage = pg.image.load('images/robot2.bmp')
self.image = pg.image.load('images/robot3.bmp')
self.baseImage = pg.image.load('images/robot3.bmp')
#self.image.fill(YELLOW)
self.image = pg.transform.scale(self.image, (TILESIZE, TILESIZE))
self.baseImage = pg.transform.scale(self.image, (TILESIZE, TILESIZE))
@ -22,6 +23,8 @@ class Player(pg.sprite.Sprite):
self.direction = direction
self.maze = Maze()
self.moves = ''
self.my_learning = Learning()
self.decision_tree_learning()
def set_direction(self, direction):
self.direction = direction
@ -88,6 +91,7 @@ class Player(pg.sprite.Sprite):
print("I move: " + str(self.direction))
print("My direction is: " + str(self.direction))
self.check_bomb()
def check_border(self, dx=0, dy=0):
@ -180,6 +184,32 @@ class Player(pg.sprite.Sprite):
print(self.moves)
def decision_tree_learning(self):
self.my_learning.load_data()
self.my_learning.learn()
self.my_learning.draw_tree()
#my_learning.predict()
""" sprawdzenie danych miny """
def check_bomb(self):
if self.check_if_on_mine():
current_mine = self.get_my_mine_object()
mine_params = current_mine.get_parameters()
self.my_learning.predict(mine_params)
return
def check_if_on_mine(self):
for mine in self.game.mines:
if mine.x == self.x and mine.y == self.y:
return True
return False
def get_my_mine_object(self):
if self.check_if_on_mine():
for mine in self.game.mines:
if mine.x == self.x and mine.y == self.y:
return mine
@ -215,6 +245,21 @@ class Mine(pg.sprite.Sprite):
self.x = x
self.y = y
def set_parameters(self, glebokosc, rozmiar, masa, moc, szkodliwosc, zabawka, teledysk, stan):
self.glebokosc = glebokosc
self.rozmiar = rozmiar
self.masa = masa
self.moc = moc
self.szkodliwosc = szkodliwosc
self.zabawka = zabawka
self.teledysk = teledysk
self.stan = stan
def get_parameters(self):
param_array = [self.glebokosc, self.rozmiar, self.masa, self.moc, self.szkodliwosc, self.zabawka, self.teledysk, self.stan]
return param_array
def update(self):
self.rect.x = self.x * TILESIZE
self.rect.y = self.y * TILESIZE