def setUp(self):
self.red = NeuralNetwork(2, [1, 1])
self.layer = NeuralLayer(2, 1)
self.layer2 = NeuralLayer(1, 1)
self.neuron = Neuron(2)
self.neuron2 = Neuron(1)
self.red2 = NeuralNetwork(2, [2, 2])
self.lay1 = NeuralLayer(2, 2)
self.lay2 = NeuralLayer(2, 2)
self.neur1 = Neuron(2)
self.neur2 = Neuron(2)
self.neur3 = Neuron(2)
self.neur4 = Neuron(2)
def __init__(self, nEpochs=500, nInputs=8, nNeurCapas=[8, 19]):
self.red = NeuralNetwork(nInputs, nNeurCapas)
self.lineList = trainReader()
self.outputList = []
shuffle(self.lineList)
self.outlist(self.lineList, self.outputList)
# self.datatrain(nEpochs)
print("Funciono Entrenamiento: CheckPoint 1")
self.dataList = dataReader()
self.dataOut = []
self.outlist(self.dataList, self.dataOut)
self.curveTraining()
def learn_network(self):
count_neurons1 = self.count_neurons_l1_spinbox.value()
count_neurons2 = self.count_neurons_l2_spinbox.value()
count_inputs_n1 = self.neuron_l1_count_inputs_spinbox.value()
count_component_input_n1 = self.neuron_l1_count_input_components_spinbox.value(
)
count_inputs_n2 = self.neuron_l2_count_inputs_spinbox.value()
self.ours = FileReader.read_matrix_files(self.ours_paths,
count_inputs_n1,
count_component_input_n1)
self.aliens = FileReader.read_matrix_files(self.aliens_paths,
count_inputs_n1,
count_component_input_n1)
self.network = nn.NeuralNetwork(h=count_inputs_n1,
g=count_inputs_n2,
components=count_component_input_n1,
n1=count_neurons1,
n2=count_neurons2)
self.network.learn(self.k1, self.k2, self.ours, self.aliens)
QMessageBox.about(None, "Learn neural network", "Done!")
def __init__(self, nTrain=1000, nInputs=2, nNeurCapas=(2, 3, 1)):
self.Network = NeuralNetwork(nInputs, nNeurCapas)
self.Inputs = []
self.Outputs = []
I1 = [0, 0]
I2 = [0, 1]
I3 = [1, 0]
I4 = [1, 1]
for i in range(nTrain):
self.Inputs.append(I1)
self.Inputs.append(I2)
self.Inputs.append(I3)
self.Inputs.append(I4)
self.Outputs.append([xor(I1[0], I1[1])])
self.Outputs.append([xor(I2[0], I2[1])])
self.Outputs.append([xor(I3[0], I3[1])])
self.Outputs.append([xor(I4[0], I4[1])])
self.Network.realTraining(self.Inputs, self.Outputs)
self.Network.feed([0, 0])
print(self.Network.getOutput())
def __init__(self, name, map):
super(Animal, self).__init__(name)
self.map = map
self.brain = NeuralNetwork(10, 9, 5)
self.x_pos = random.randint(0, self.map.MAPWIDTH - 1)
self.y_pos = random.randint(0, self.map.MAPHEIGHT - 1)
self.hunger = 0
self.tiredness = 0
self.distance_x = 0
self.distance_y = 0
self.fitness = 0
self.rect = self.image.get_rect()
def load_network(self):
fileName = self.__openFileNameDialog("Network config file (*.json)")
if fileName:
with open(fileName, 'r') as file_in:
file_text = file_in.read()
json_obj = json.loads(file_text)
json_obj['mu'] = np.array(json_obj['mu'])
json_obj['W'] = np.array(json_obj['W'])
self.network = nn.NeuralNetwork()
self.network.__dict__ = json_obj
QMessageBox.about(None, "Loading neural network", "Done!")
from Network import NeuralNetwork
import random
import numpy as np
import matplotlib.pyplot as plt
import cv2
input_nodes = 9
hidden_nodes = 5
output_nodes = 2
learning_rate = 0.2
n = NeuralNetwork(input_nodes, hidden_nodes, output_nodes, learning_rate)
ppi = np.array([[0.3, 0.3]])
epochs = 1111
for i in range(epochs):
a = np.random.permutation(18)
for pro in a:
number = pro
# number = random.randint(0, 17)
if number == 1 or number == 2 or number == 3:
rightAnswers = np.array([[1, 0]])
elif number == 4 or number == 5 or number == 6:
rightAnswers = np.array([[0, 1]])
elif 6 < number < 16 or number == 17:
rightAnswers = np.array([[1, 1]])
elif number == 0 or number == 16:
rightAnswers = np.array([[0, 0]])
class dataSet:
def __init__(self, nEpochs=500, nInputs=8, nNeurCapas=[8, 19]):
self.red = NeuralNetwork(nInputs, nNeurCapas)
self.lineList = trainReader()
self.outputList = []
shuffle(self.lineList)
self.outlist(self.lineList, self.outputList)
# self.datatrain(nEpochs)
print("Funciono Entrenamiento: CheckPoint 1")
self.dataList = dataReader()
self.dataOut = []
self.outlist(self.dataList, self.dataOut)
self.curveTraining()
def outlist(self, lineList, outList):
for i in lineList:
zerolist = [0] * 19
zerolist[i.pop(0)] = 1
outList.append(zerolist)
def datatrain(self, nEpoch):
for j in range(nEpoch):
#start = time.time()
self.red.realTraining(self.lineList, self.outputList)
#end = time.time()
#print(end - start)
def curveTraining(self):
totallineas = len(self.dataList)
contpres = 0.0
error = 0
xaprendizaje = []
yaprendizaje = []
xerror = []
yerror = []
for k in range(7):
for l in range(len(self.dataList)):
self.red.feed(self.dataList[l])
for m in range(len(self.red.output)):
absdif = abs(self.dataOut[l][m] - self.red.output[m])
if absdif < 0.5:
contpres += 1
error += absdif**2
contpres /= 19
contpres /= totallineas
xaprendizaje.append(k)
yaprendizaje.append(contpres)
contpres = 0
xerror.append(k)
yerror.append(error)
error = 0
self.datatrain(k * 100 + 100)
plt.figure(1)
plt.subplot(211)
plt.xlabel("Numero de Epochs")
plt.ylabel("Cura de Aprendizaje")
plt.plot(xaprendizaje, yaprendizaje, 'b-')
plt.subplot(212)
plt.xlabel("Numero de Epochs")
plt.ylabel("Error")
plt.plot(xerror, yerror, 'r-')
plt.show()
class NetworkTests(unittest.TestCase):
def setUp(self):
self.red = NeuralNetwork(2, [1, 1])
self.layer = NeuralLayer(2, 1)
self.layer2 = NeuralLayer(1, 1)
self.neuron = Neuron(2)
self.neuron2 = Neuron(1)
self.red2 = NeuralNetwork(2, [2, 2])
self.lay1 = NeuralLayer(2, 2)
self.lay2 = NeuralLayer(2, 2)
self.neur1 = Neuron(2)
self.neur2 = Neuron(2)
self.neur3 = Neuron(2)
self.neur4 = Neuron(2)
# testcase entregado en ucursos
def testCase1(self):
self.neuron.custom([0.4, 0.3], 0.5)
self.neuron2.custom([0.3], 0.4)
self.layer.custom([self.neuron])
self.layer2.custom([self.neuron2])
self.red.custom([self.layer, self.layer2])
self.red.train([1, 1], [1])
self.assertEqual(round(self.neuron.weight[0], 15), 0.402101508999489)
self.assertEqual(round(self.neuron.weight[1], 15), 0.302101508999489)
self.assertEqual(round(self.neuron2.weight[0], 15), 0.330262548639919)
self.assertEqual(round(self.neuron.bias, 15), 0.502101508999489)
self.assertEqual(round(self.neuron2.bias, 15), 0.439377453127974)
def testCase2(self):
self.neur1.custom([0.7, 0.3], 0.5)
self.neur2.custom([0.3, 0.7], 0.4)
self.neur3.custom([0.2, 0.3], 0.3)
self.neur4.custom([0.4, 0.2], 0.6)
self.lay1.custom([self.neur1, self.neur2])
self.lay2.custom([self.neur3, self.neur4])
self.red2.custom([self.lay1, self.lay2])
self.red2.train([1, 1], [1, 1])
self.assertEqual(round(self.neur1.weight[0], 15), 0.702510448549328)
self.assertEqual(round(self.neur1.weight[1], 15), 0.302510448549328)
self.assertEqual(round(self.neur2.weight[0], 15), 0.302498011357483)
self.assertEqual(round(self.neur2.weight[1], 15), 0.702498011357483)
self.assertEqual(round(self.neur3.weight[0], 15), 0.229947378819557)
self.assertEqual(round(self.neur3.weight[1], 15), 0.329383628639501)
self.assertEqual(round(self.neur4.weight[0], 15), 0.419430056526462)
self.assertEqual(round(self.neur4.weight[1], 15), 0.219064291698386)
self.assertEqual(round(self.neur1.bias, 15), 0.502510448549328)
self.assertEqual(round(self.neur2.bias, 15), 0.402498011357483)
self.assertEqual(round(self.neur3.bias, 15), 0.336629542251590)
self.assertEqual(round(self.neur4.bias, 15), 0.623765488150905)
class Animal(Sprite):
def __init__(self, name, map):
super(Animal, self).__init__(name)
self.map = map
self.brain = NeuralNetwork(10, 9, 5)
self.x_pos = random.randint(0, self.map.MAPWIDTH - 1)
self.y_pos = random.randint(0, self.map.MAPHEIGHT - 1)
self.hunger = 0
self.tiredness = 0
self.distance_x = 0
self.distance_y = 0
self.fitness = 0
self.rect = self.image.get_rect()
# Think Outputs
#--------------------------------------------------------
def move_up(self):
if self.valid_move("up"):
self.y_pos -= 1
self.eat()
def move_down(self):
if self.valid_move("down"):
self.y_pos += 1
self.eat()
def move_right(self):
if self.valid_move("right"):
self.x_pos += 1
self.eat()
def move_left(self):
if self.valid_move("left"):
self.x_pos -= 1
self.eat()
def sleep(self):
self.tiredness -= 10
self.hunger += 5
def valid_move(self, dir):
if self.y_pos >= 0 and self.y_pos < (self.map.MAPHEIGHT - 1) \
and self.x_pos >= 0 and self.x_pos < (self.map.MAPWIDTH - 1):
if (dir == "left" and self.map.Tiles[self.y_pos][self.x_pos - 1].__str__() == 'GRASS')\
or (dir == "right" and self.map.Tiles[self.y_pos][self.x_pos + 1].__str__() == 'GRASS')\
or (dir == "down" and self.map.Tiles[self.y_pos + 1][self.x_pos].__str__() == 'GRASS')\
or (dir == "up" and self.map.Tiles[self.y_pos - 1][self.x_pos].__str__() == 'GRASS'):
return True
return False
def eat(self):
if self.map.Tiles[self.y_pos][self.x_pos].is_occupied:
self.hunger -= 20
self.fitness += 25
p = self.map.plants.sprites()
for eaten in p:
if eaten.x_pos == self.x_pos and eaten.y_pos == self.y_pos:
self.map.plants.remove(eaten)
# Think & Helper Functions
# --------------------------------------------------------------------
def think(self):
self.tiredness += 5
self.hunger += 5
self.update_fitness()
self.find_food()
inputs = self.make_inputs()
output = self.brain.feedForward(inputs)
decision = output.index(max(output))
if decision == 0:
self.move_up()
elif decision == 1:
self.move_down()
elif decision == 2:
self.move_right()
elif decision == 3:
self.move_left()
elif decision == 4:
self.sleep()
def make_inputs(self):
inputs = [self.x_pos, self.y_pos, self.hunger, self.tiredness, self.distance_x, self.distance_y]
if self.y_pos >= 0 and self.y_pos < (self.map.MAPHEIGHT - 1) \
and self.x_pos >= 0 and self.x_pos < (self.map.MAPWIDTH - 1):
obstruction_left = (0 if self.map.Tiles[self.y_pos][self.x_pos - 1].__str__() == 'GRASS' else 1)
obstruction_right = (0 if self.map.Tiles[self.y_pos][self.x_pos + 1].__str__() == 'GRASS' else 1)
obstruction_up = (0 if self.map.Tiles[self.y_pos + 1][self.x_pos].__str__() == 'GRASS' else 1)
obstruction_down = (0 if self.map.Tiles[self.y_pos - 1][self.x_pos].__str__() == 'GRASS' else 1)
else:
obstruction_left = (1 if self.x_pos == 0 else 0)
obstruction_right = (1 if self.x_pos == self.map.MAPWIDTH else 0)
obstruction_up = (1 if self.y_pos == 0 else 0)
obstruction_down = (1 if self.x_pos == self.map.MAPHEIGHT else 0)
inputs.append(obstruction_left)
inputs.append(obstruction_right)
inputs.append(obstruction_up)
inputs.append(obstruction_down)
return inputs
def update_fitness(self):
if self.hunger < 25:
self.fitness += 2
elif self.hunger < 50:
self.fitness += 1
if self.tiredness < 25:
self.fitness += 2
elif self.tiredness < 50:
self.fitness += 1
if self.hunger & self.tiredness < 25:
self.fitness += 4
elif self.hunger & self.tiredness < 50:
self.fitness += 2
print("fitness:" + self.fitness.__str__())
def find_food(self):
current_distance = - 1
closest_x = -1
closest_y = -1
for plant in self.map.plants:
plant_x = plant.x_pos
plant_y = plant.y_pos
x_dist = plant_x - self.x_pos
y_dist = plant_y - self.y_pos
distance = math.sqrt((x_dist * x_dist) + (y_dist * y_dist))
if distance < current_distance or current_distance == -1:
closest_x = plant_x
closest_y = plant_y
self.distance_x = closest_x
self.distance_y = closest_y
def valid_fauna(self):
if self.map.Tiles[self.y_pos][self.x_pos].__str__() == 'GRASS'\
and self.map.Tiles[self.y_pos][self.x_pos].is_occupied == False:
return True
else:
return False
def main():
# Creating NN
nn = NeuralNetwork([784, 64, 3])
# Creating training data
createTraining(cats_training, "cat")
createTraining(trains_training, "train")
createTraining(rainbows_training, "rainbow")
# Creating test data
createTest(cats_testing, "cat")
createTest(trains_testing, "train")
createTest(rainbows_testing, "rainbow")
# make one array from all the training and test data
training_data = cats_training + trains_training + rainbows_training
testing_data = cats_testing + trains_testing + rainbows_testing
# Normalize the training data and testing data, (0-1)
normalizeValues(trains_training)
normalizeValues(testing_data)
# arrange the data in specific format that will fit to the NN library i created.
for j in range(2400):
for i in range(785):
training_data[j][i] = [training_data[j][i]]
for j in range(600):
for i in range(785):
testing_data[j][i] = [testing_data[j][i]]
'''
training the model with the training data (30 times),
shuffle the training data every iteration.
specific target to every label.
'''
for i in range(35):
random.shuffle(training_data)
for j in range(2400):
target = [[0], [0], [0]]
if (training_data[j][784] == ['cat']):
target[0] = [1]
nn.train(training_data[j][:-1], target, 1)
if (training_data[j][784] == ['train']):
target[1] = [1]
nn.train(training_data[j][:-1], target, 1)
if (training_data[j][784] == ['rainbow']):
target[2] = [1]
nn.train(training_data[j][:-1], target, 1)
'''
Calculate the acc. should be more various of training data.
currently : 800 cats, 800 rainbows, 800 trains.
'''
acc = 0
trains = 0
rainbows = 0
cats = 0
for i in range(len(testing_data)):
label = testing_data[i][784]
guess = nn.feed_forword(testing_data[i][:-1])
if max(guess) == guess[0] and label == ['cat']:
acc += 1
cats += 1
if max(guess) == guess[1] and label == ['train']:
acc += 1
trains += 1
if max(guess) == guess[2] and label == ['rainbow']:
acc += 1
rainbows += 1
print("Acc : " + str(acc))
print("length of testing data : " + str(len(testing_data)))
print("Cats : " + str(cats))
print("Rainbows : " + str(rainbows))
print("trains : " + str(trains))
from Network import NeuralNetwork
import cv2
import numpy as np
import matplotlib.pyplot as plt
n = NeuralNetwork(9, 5, 2, 0.3)
n.synaptic_weight1 = np.load('./weights/example_1.npy')
n.synaptic_weight2 = np.load('./weights/example_2.npy')
image = cv2.imread("dataset/test/test2.0.png", 0)
inputs = image.flatten()
inputs = np.asfarray(inputs / 255.0 * 0.99) + 0.01
answers = n.query(inputs)
print(str(answers[0]) + str(answers[1]))
plt.imshow(image)
plt.xlabel('test2.0.png')
plt.title(str(answers[0]) + ' -- ' + str(answers[1]))
plt.show()
for i in range(18):
image = cv2.imread("dataset/test/" + str(i) + ".png", 0)
inputs = image.flatten()
inputs = np.asfarray(inputs / 255.0 * 0.99) + 0.01
answers = n.query(inputs)
print(str(answers[0]) + str(answers[1]))
