from keras.layers import Dense
display_width = 500
display_height = 500
green = (0, 255, 0)
black = (0, 0, 0)
red = (255, 0, 0)
white = (255, 255, 255)
pygame.init()
display = pygame.display.set_mode((display_width, display_height))
clock = pygame.time.Clock()
#Left is 0, right is 1, down is 2, up is 3
training_data_x, training_data_y = generate_training_data(display, clock)
model = Sequential()
model.add(Dense(units=9, input_dim=7))
model.add(Dense(units=15, activation='relu'))
model.add(Dense(activation="softmax", units=3))
model.compile(loss='mean_squared_error',
optimizer='adam',
metrics=['accuracy'])
model.fit((np.array(training_data_x).reshape(-1, 7)),
(np.array(training_data_y).reshape(-1, 3)),
batch_size=256,
epochs=10)
from keras.models import Sequential from training_data import generate_training_data import numpy from keras.utils import * from keras.layers import Dense numpy.set_printoptions(threshold=numpy.nan) model = Sequential() #make model with two hidden layers of 21 neurons and a single output neuron model.add(Dense(units=7, activation='relu', input_dim=7)) model.add(Dense(units=7, activation='relu')) model.add(Dense(units=7, activation='sigmoid')) model.add(Dense(units=7, activation='relu')) model.add(Dense(units=1, activation='sigmoid')) x_train, y_train = generate_training_data() model.compile(loss='mean_squared_error', optimizer='rmsprop', metrics=['accuracy']) model.fit(x_train, y_train, epochs=5, batch_size=10) print model.predict_classes(numpy.array([[0.0,0.0,1.0,1.0,1.0,0.0,0.0]]))