def plant_seed(self):
# Function for planting trees
# Test to see if the tile the arborist is on is a tile that a tree can be planted on
if TileFuncs.get_tile(self.arborist.world,self.arborist.location).plantable == 1:
self.arborist.hit = 0
self.arborist.update()
old_tile = TileFuncs.get_tile(self.arborist.world,Vector2(self.arborist.location))
darkness = pygame.Surface((32, 32))
darkness.set_alpha(old_tile.darkness)
new_tile = Tile.Baby_Tree(self.arborist.world, "GrassWithCenterTree")
new_tile.darkness = old_tile.darkness
new_tile.location = TileFuncs.get_tile_pos(self.arborist.world,self.arborist.destination)*32
new_tile.rect.topleft = new_tile.location
new_tile.color = old_tile.color
self.arborist.world.tile_array[int(new_tile.location.y/32)][int(new_tile.location.x/32)] = new_tile
self.arborist.world.world_surface.blit(new_tile.img, new_tile.location)
self.arborist.world.world_surface.blit(darkness, new_tile.location)
# Goes to a random destination no matter what
self.arborist.hit = 0
BaseFunctions.random_dest(self.arborist)
def do_actions(self):
if self.arborist.location == self.arborist.destination and self.arborist.hit >= 4 and TileFuncs.get_tile(
self.arborist.world,self.arborist.location).plantable == 1:
self.plant_seed()
if self.arborist.location == self.arborist.destination and self.arborist.hit != 4 and TileFuncs.get_tile(
self.arborist.world, self.arborist.location).plantable != 1:
BaseFunctions.random_dest(self.arborist)
def check_conditions(self):
if self.angler.location.get_distance_to(self.angler.destination) < self.angler.max_speed:
location_array = TileFuncs.get_vnn_array(self.angler.world,(self.angler.location), self.angler.view_range)
for location in location_array:
# TODO: This will make the angler go into the water, change this to go to the nearest walkable tile.
test_tile = TileFuncs.get_tile(self.angler.world, location)
if test_tile.__class__.__name__ == "WaterTile":
self.angler.destination = location.copy()
return "Fishing"
BaseFunctions.random_dest(self.angler)
def check_conditions(self):
if self.lumberjack.location.get_distance_to(self.lumberjack.destination) < 15:
location_array = TileFuncs.get_vnn_array(self.lumberjack.world,(self.lumberjack.location), self.lumberjack.view_range)
for location in location_array:
test_tile = TileFuncs.get_tile(self.lumberjack.world,location)
if test_tile.name == "GrassWithCenterTree":
self.lumberjack.Tree_tile = test_tile
self.lumberjack.tree_id = test_tile.id
self.lumberjack.destination = location.copy()
return "Chopping"
BaseFunctions.random_dest(self.lumberjack)
def check_conditions(self):
if self.angler.location.get_distance_to(
self.angler.destination) < self.angler.max_speed:
location_array = TileFuncs.get_vnn_array(self.angler.world,
(self.angler.location),
self.angler.view_range)
for location in location_array:
# TODO: This will make the angler go into the water, change this to go to the nearest walkable tile.
test_tile = TileFuncs.get_tile(self.angler.world, location)
if test_tile.__class__.__name__ == "WaterTile":
self.angler.destination = location.copy()
return "Fishing"
BaseFunctions.random_dest(self.angler)
def check_conditions(self):
if self.lumberjack.location.get_distance_to(
self.lumberjack.destination) < 15:
location_array = TileFuncs.get_vnn_array(
self.lumberjack.world, (self.lumberjack.location),
self.lumberjack.view_range)
for location in location_array:
test_tile = TileFuncs.get_tile(self.lumberjack.world, location)
if test_tile.name == "GrassWithCenterTree":
self.lumberjack.Tree_tile = test_tile
self.lumberjack.tree_id = test_tile.id
self.lumberjack.destination = location.copy()
return "Chopping"
BaseFunctions.random_dest(self.lumberjack)
def feedFoward(self,X):
a = X.copy() # To avoid changing the values of the input outside the scope of the method.
aList = []
for w,b in zip(self.weights,self.bias):
z = numpy.add(a@w,b) # Applying layer weight to previous activation and adding bias.
a = BaseFunctions.sigmoid(z) # Activating z with the sigmoid function.
aList.append(a)
return aList
def do_actions(self):
current_tile = TileFuncs.get_tile(self.farmer.world, self.farmer.location)
if current_tile.tillable:
darkness = pygame.Surface((self.farmer.TileSize, self.farmer.TileSize))
darkness.set_alpha(current_tile.darkness)
new_tile = Tile.SoilTile(self.farmer.world, "Soil2")
new_tile.darkness = darkness
new_tile.location = current_tile.location
new_tile.rect.topleft = new_tile.location
new_tile.color = current_tile.color # TODO: Figure out what this does.
self.farmer.world.tile_array[int(new_tile.location.y / 32)][int(new_tile.location.x / 32)] = new_tile
self.farmer.world.world_surface.blit(new_tile.img, new_tile.location)
self.farmer.world.world_surface.blit(darkness, new_tile.location)
# TODO: Update the minimap
BaseFunctions.random_dest(self.farmer)
elif self.farmer.location.get_distance_to(self.farmer.destination) < self.farmer.speed:
BaseFunctions.random_dest(self.farmer)
def check_conditions(self):
"""Check if the explorer should still be exploring"""
if self.explorer.location.get_distance_to(self.explorer.destination) <= self.explorer.speed:
BaseFunctions.random_dest(self.explorer)
def entry_actions(self):
"""When the explorer starts exploring (enters exploring state)."""
BaseFunctions.random_dest(self.explorer)
def entry_actions(self):
BaseFunctions.random_dest(self.lumberjack)
# Training the data
#
# the traingset will have 130000 examples, since it is too heavy to load a 130000*784 array in one variable the data
# will flow throught batches. 130 batches with 1000 examples each.
for i in range(130):
X = []
Y = []
for j in range(1000):
TrainingSample = list(map(int, f.readline().split()))
# the first element of each row is the answer of the case
# it will be converged to an array in order to be compared to the output of the neural network.
y = BaseFunctions.intToRow(TrainingSample[0], 10)
Y.append(y)
x = TrainingSample[1:]
X.append(x)
# to avoid overflow, the scale of X's values will be reduced.
# now each element of x will be in the range [0,1]
# 0 -> pure black
# 1 -> pure white
X = numpy.array(X) / 255
Y = numpy.array(Y)
C = Nu.Cost(X, Y)
print("Cost of Batch {} = {}".format(i + 1, C))
def predict(self,X):
a = self.feedFoward(X)[-1]
a = BaseFunctions.moldData(a) # rouding the values.
return a
def entry_actions(self):
BaseFunctions.random_dest(self.angler)
def entry_actions(self):
BaseFunctions.random_dest(self.angler)
import os
os.chdir("/Users/MRisk/Desktop/SpaceFillingCurves/SpaceFilling")
import BaseFunctions as SFC
import numpy as np
import matplotlib.pyplot as plt
plotA = plt.figure(figsize=(7, 7))
plotA = plt.plot(*zip(*SFC.Hilbert_Polygon(5)), color="black", linewidth=5)
plotA = plt.axis('off', emit=False)
plt.show()
x1, y1 = zip(*SFC.Hilbert_Polygon(2))
x2, y2 = zip(*SFC.Hilbert_Polygon(3))
fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(6, 3))
axes[0].plot(x1, y1, linewidth=3, color="black")
axes[1].plot(x2, y2, linewidth=3, color="black")
axes[0].grid(b=True)
axes[1].grid(b=True)
axes[0].set_xticks([0, 0.25, 0.5, 0.75, 1])
axes[0].set_yticks([0, 0.25, 0.5, 0.75, 1])
axes[1].set_xticks([0, 0.25, 0.5, 0.75, 1])
axes[1].set_yticks([0, 0.25, 0.5, 0.75, 1])
axes[0].set_xticklabels([0, "", "", "", 1])
axes[0].set_yticklabels([0, "", "", "", 1])
axes[1].set_xticklabels([0, "", "", "", 1])
axes[1].set_yticklabels([0, "", "", "", 1])
axes[0].set_xlim([0, 1])
axes[0].set_ylim([0, 1])
axes[1].set_xlim([0, 1])
axes[1].set_ylim([0, 1])
def entry_actions(self):
BaseFunctions.random_dest(self.lumberjack)
def check_conditions(self):
if self.fishing_ship.location.get_distance_to(self.fishing_ship.destination) < self.fishing_ship.speed:
BaseFunctions.random_dest(self.fishing_ship)
def entry_actions(self):
BaseFunctions.random_dest(self.arborist)
def entry_actions(self):
BaseFunctions.random_dest(self.farmer)