def test_xor():
"""
This method will run the test on xor dataset.
"""
dg = xor() # Initialize a dataset creator
training_data, training_labels = dg.query_data(samples=100)
n = xor_net(training_data,
training_labels) # This call should return a net object
params = n.get_params(
) # This call should reaturn parameters of the model that are
# fully trained.
testing_data, testing_labels = dg.query_data(
samples=100) # Create a random testing dataset.
predictions = n.get_predictions(
testing_data) # This call should return predictions.
print "Accuracy of predictions on XOR data = " + str(
accuracy(testing_labels, predictions)) + "%"
import sys
sys.path.append('../')
from tools.trainer import trainer, poly_trainer
from network import expert, novice, judge
from dataset import xor
from globals import *
if __name__ == '__main__':
dataset = xor()
################ Expert ################
print(" \n\n Expert Assembly \n\n")
expert_net = expert(images=dataset.images)
expert_net.cook(labels=dataset.labels)
expert_bp = trainer(expert_net,
dataset.feed,
init_vars=False,
tensorboard='expert')
################ Indpendent Novice ################
"""
print (" \n\n Independent Novice Assembly \n\n")
indep_net = novice(images = dataset.images,
name = 'novice_independent')
indep_net.cook( labels = dataset.labels )
indep_bp = trainer( indep_net,
session = expert_bp.session,
dataset.feed)
"""
num_classes = 26
(X, y) = dataset.character()
hidden_layer = 20
data = np.array([
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1,
0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 0,
0, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0
]).reshape(1, 64)
elif (args["type"] == '2'):
num_classes = 6
(X, y) = dataset.lima_bit()
hidden_layer = 12
data = np.array([0, 0, 0, 1, 0]).reshape(1, 5)
elif (args["type"] == '3'):
num_classes = 2
(X, y) = dataset.xor()
hidden_layer = 2
data = np.array([0, 0]).reshape(1, 2)
elif (args["type"] == '4'):
num_classes = 2
(X, y) = dataset.tiga_bit()
hidden_layer = 3
data = np.array([0, 1, 0]).reshape(1, 3)
num_inputs = X.shape[1]
shape = (num_inputs, hidden_layer, num_classes)
# Set up
cost_func = functools.partial(eval_neural_network, shape=shape, X=X, y=y)
swarm = pso.ParticleSwarm(cost_func, dim=dim_weights(shape), size=50)