Skip to content

GitLab

Commit 1cc8c5ce authored by Antoine Roux's avatar Antoine Roux
Browse files

added genetic, works so-so : achieves 1024 from time to time

parent 0486ffd9
Hide whitespace changes
Inline Side-by-side
2048.py
View file @ 1cc8c5ce
......@@ -2,6 +2,7 @@
import numpy as np
import random
import time
import copy
import matplotlib.pyplot as plt
##CMA-ES
......@@ -43,7 +44,7 @@ class Grid (object) :
# [16, 8, 4, 2],
# [32, 16, 8, 4],
# [64, 32, 16, 8]])
self.fitnessGrid = givenFitnessGrid
self.fitnessGrid = np.array(givenFitnessGrid)
def __str__ (self) :
string = ""
......@@ -181,12 +182,12 @@ class Grid (object) :
def calcFitness (self) :
""" Returns the fitness of the grid
"""
# fitnessGrid = [[64, 16, 4, 1], #a peu près pareil
# fitnessGrid = [[64, 16, 4, 1], #a peu pres pareil
# [256, 64, 16, 4],
# [1024, 256, 64, 16],
# [4096, 1024, 256, 64]]
## pas nécessairement plus lent...
## pas necessairement plus lent...
# fitness = 0
# for k in range(4) :
# for i in range (4) :
......@@ -197,7 +198,7 @@ class Grid (object) :
class Generation (object) :
def __init__ (self, givenSizeOfPopulation) :
def __init__ (self, givenSizeOfPopulation, initValue) :
"""
"""
self.sizeOfPopulation = givenSizeOfPopulation
......@@ -211,7 +212,7 @@ class Generation (object) :
[0, 0, 0, 0]])
for i in range (4) :
for j in range (4) :
fitnessGrid[i,j] = random.randint(0, 10)
fitnessGrid[i,j] = random.randint(0, initValue)
self.individuals.append(Grid([[0, 0, 0, 0],
[0, 0, 0, 0],
......@@ -234,7 +235,6 @@ class Generation (object) :
for k in range(self.sizeOfPopulation) :
currentFitness = 0
for i in range(nbrIterations) :
print("k, i =", k, i)
finishedGrid = single_AI(self.individuals[k].grid, self.individuals[k].fitnessGrid)
currentFitness += sum(sum(finishedGrid))
self.fitnesses.append(int(currentFitness / nbrIterations))
......@@ -250,10 +250,12 @@ class Generation (object) :
nbrOfOthers = int((self.sizeOfPopulation) * proportionOfOthers)
indexes = []
listOfFitnesses = copy.deepcopy(self.fitnesses)
for k in range(nbrOfBest) :
currentMaxIndex = self.fitnesses.index(max(self.fitnesses))
currentMaxIndex = listOfFitnesses.index(max(listOfFitnesses))
indexes.append(currentMaxIndex)
self.fitnesses[currentMaxIndex] = 0
listOfFitnesses[currentMaxIndex] = 0
for k in range (nbrOfOthers) :
randomlySelected = random.randint(0, self.sizeOfPopulation-1)
......@@ -261,8 +263,6 @@ class Generation (object) :
randomlySelected = random.randint(0, self.sizeOfPopulation-1)
indexes.append(randomlySelected)
self.fitnesses = [] #we reset the fitnesses
#we delete the individuals not selected
for k in range(self.sizeOfPopulation) :
if k not in indexes :
......@@ -304,17 +304,17 @@ class Generation (object) :
parent2 = random.choice(indexes)
self.individuals[k] = Grid([[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0]] ,
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0]] ,
[[0, 0, 0, 0],
[[1, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0]])
for i in range(4) :
for j in range(4) :
self.individuals[k].fitnessGrid[i][j] = 0.5 * (self.individuals[parent1].grid[i,j] + self.individuals[parent2].grid[i,j])
self.individuals[k].fitnessGrid[i][j] = 0.5 * (self.individuals[parent1].fitnessGrid[i,j] + self.individuals[parent2].fitnessGrid[i,j])
......@@ -344,7 +344,7 @@ def single_AI(givenGrid, givenFitnessGrid) :
myGrid.addNbr()
while True :
fitnessGrid = np.array([[0, 0, 0, 0], #lignes = à 1er move constant
fitnessGrid = np.array([[0, 0, 0, 0], #lignes = a 1er move constant
[0, 0, 0, 0],
[0, 0, 0, 0],
[0, 0, 0, 0]])
......@@ -356,10 +356,12 @@ def single_AI(givenGrid, givenFitnessGrid) :
if sum(possibleMoves) != 0 : #if at least one move is possible
consideredMovesArray = [0, 1, 2, 3]
if (possibleMoves[0] == True or possibleMoves[2] == True or possibleMoves[3] == True) :
#if can swipe up, left or down, do not swipe right
consideredMovesArray = [0, 2, 3]
# if np.count_nonzero(myGrid.grid[:,0]) != 0 :
# #if first line is not full, do not swipe up
# #PAS SUFFISANT CAR MM SI LA LIGNE EST PLEINE SWIPER PEUT AVOIR UN EFFET
......@@ -452,7 +454,7 @@ if MODE == "PLAY" :
# finishedGrid = single_AI()
# endTime = time.time()
# print(" \n----- Partie finie -----")
# print("Temps écoulé :", int((endTime-startTime)*1000), "ms")
# print("Temps ecoule :", int((endTime-startTime)*1000), "ms")
# print("Valeur max :", evaluateGrid(finishedGrid))
# print("Score brut :", sum(sum(finishedGrid)))
# print(Grid(finishedGrid))
......@@ -475,9 +477,9 @@ if MODE == "PLAY" :
# print("Partie", k+1, "sur", nbrGames, " - Score brut :", rawScore)
# endTime = time.time()
# print(" \n----- Parties finies -----")
# print("Temps écoulé :", int((endTime-startTime)*1000), "ms")
# print("Temps ecoule :", int((endTime-startTime)*1000), "ms")
# print(np.array(listScores))
# print("Temps par unité :", (endTime-startTime)/sumRawScores*1000, "ms")
# print("Temps par unite :", (endTime-startTime)/sumRawScores*1000, "ms")
# print("--------------------------")
# #plot a bar graph
# width = 30
......@@ -485,11 +487,39 @@ if MODE == "PLAY" :
# plt.show()
if MODE == "GENETIC" :
nbrOfIndividuals = 30
nbrOfGenerations = 15
nbrOfEvaluations = 100
initValue = 30
myGeneration = Generation(nbrOfIndividuals, initValue)
generationCounter = 0
while generationCounter < nbrOfGenerations :
generationCounter += 1
print(generationCounter)
if generationCounter < 5 :
selectionRate = 0.5
elif generationCounter < 10:
selectionRate = 0.35
else :
selectionRate = 0.2
myGeneration.evaluate(nbrOfEvaluations)
indexes = myGeneration.select(selectionRate, 0)
print(indexes)
print(sum(myGeneration.fitnesses)/nbrOfIndividuals)
myGeneration.fitnesses = []
myGeneration.reproduce(indexes)
text_file = open("Output.txt", "w")
string = ""
for k in range(nbrOfIndividuals) :
for i in range(4) :
for j in range(4) :
string += str(myGeneration.individuals[k].fitnessGrid[i][j]) + " "
string += "\n"
string += "\n"
text_file.write(string)
myGeneration = Generation(10)
myGeneration.evaluate(10)
indexes = myGeneration.select(0.3, 0)
print(indexes)
print(myGeneration)
myGeneration. reproduce(indexes)
print(myGeneration)
text_file.close()
2048_ma_fitness.py
View file @ 1cc8c5ce
......@@ -159,15 +159,22 @@ class Grid (object) :
""" Returns the fitness of the grid
"""
# fitnessArray = [[64, 16, 4, 1], #a peu près pareil
# [256, 64, 16, 4],
# [1024, 256, 64, 16],
# [4096, 1024, 256, 64]]
fitnessArray = np.array([[8, 4, 2, 1],
[16, 8, 4, 2],
[32, 16, 8, 4],
[64, 32, 16, 8]])
[16, 8, 4, 2],
[32, 16, 8, 4],
[64, 32, 16, 8]])
# fitnessArray = np.array([[3, 5, 6, 2], #celui du test 2
# [1, 2, 7, 5],
# [2, 3, 6, 6],
# [5, 5, 4, 1]])
# fitnessArray = np.array([[11, 20, 6, 21], #celui du test sur les ordis de l'ensta
# [13, 4, 22, 17],
# [7, 13, 23, 25],
# [12, 5, 23, 9]])
## pas nécessairement plus lent...
# fitness = 0
......@@ -235,7 +242,7 @@ def single_AI() :
# if np.count_nonzero(myGrid.grid[:,0]) != 0 :
# #if first line is not full, do not swipe up
# #PAS SUFFISANT CAR MM SI LA LIGNE EST PLEINE SWIPER PEUT AVOIR UN EFFET
# #EN PLUS CA DIMINUE LE MEILLEUR SCORE...
# #EN PLUS CA DIMINUE LE MEILLEUR SCORE car il fait alors un pur escalier
# consideredMovesArray = [2, 3]
for k in consideredMovesArray :
......@@ -320,17 +327,17 @@ if MODE == "PLAY" :
if MODE == "AI" :
startTime = time.time()
finishedGrid = single_AI()
[finishedGrid, rawScore] = single_AI()
endTime = time.time()
print(" \n----- Partie finie -----")
print("Temps écoulé :", int((endTime-startTime)*1000), "ms")
print("Valeur max :", evaluateGrid(finishedGrid))
# print("Valeur max :", evaluateGrid(finishedGrid))
print(Grid(finishedGrid))
print("------------------------")
if MODE == "MULTI_AI" :
startTime = time.time()
nbrGames = 10
nbrGames = 200
listScores = [[],[]]
sumRawScores = 0
for k in range(nbrGames) :
......
Idées 2048.pages
View file @ 1cc8c5ce
No preview for this file type
Outputs/Output2.txt 0 → 100644
View file @ 1cc8c5ce
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
3 5 6 2
1 2 7 5
2 3 6 6
5 5 4 1
Outputs/Output5.txt 0 → 100644
View file @ 1cc8c5ce
47 69 47 68
58 46 71 72
92 91 86 50
96 101 103 102
48 70 49 70
57 50 71 74
93 91 84 49
97 100 103 100
47 70 49 68
57 49 71 73
92 90 84 49
96 100 102 100
47 69 48 69
58 47 71 73
92 91 86 50
96 101 103 102
48 70 48 69
57 49 71 73
94 91 85 49
97 100 103 100
47 70 49 68
57 48 71 72
93 90 84 49
96 100 102 100
47 69 47 68
58 46 71 72
93 91 87 50
97 101 103 103
47 69 47 68
58 46 71 72
92 91 86 50
96 101 103 102
48 71 50 69
56 52 71 74
94 91 82 48
97 99 102 98
48 71 51 68
57 51 71 73
93 90 82 49
96 99 101 98
48 70 49 70
57 50 71 74
93 91 84 49
97 100 103 100
48 70 48 69
57 49 71 73
94 91 85 49
97 100 103 100
47 69 47 68
58 46 71 72
92 91 86 50
96 101 103 102
48 71 50 69
56 52 71 74
94 91 82 48
97 99 102 98
48 70 49 70
57 50 71 74
93 91 84 49
97 100 103 100
47 70 49 68
57 49 71 73
92 90 84 49
96 100 102 100
47 70 49 68
57 48 71 72
93 90 84 49
96 100 102 100
47 69 47 68
58 46 71 72
92 91 86 50
96 101 103 102
47 69 47 68
58 46 71 72
92 91 86 50
96 101 103 102
49 71 50 71
56 53 71 75
95 92 83 48
98 100 103 98
Outputs/Output6.txt 0 → 100644
View file @ 1cc8c5ce
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
7 1 3 4
3 5 6 7
2 6 7 7
6 4 4 2
Outputs/Output7.txt 0 → 100644
View file @ 1cc8c5ce
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
3 5 4 1
2 4 7 3
2 4 6 3
2 3 5 5
4 5 4 1
2 4 8 3
2 4 7 4
2 3 5 5
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment