def test_compute_classification(self):
network = RbfNetwork(2, 1, 2)
network.copy_memory([
2.0, # input 1 to RBF 1
2.0, # input 2 to RBF 1
5.0, # RBF width
2.0, # RBF, center-0
4.0, # RBF, center-1
3.0, # RBF1 to Output 1
4.0, # Bias to Output 1
5.0, # RBF1 to Output 2
6.0]) # Bias to Output 2
x = [1, 2]
y = network.compute_regression(x)
# Inputs: (2*1) + (2*2) = 6
# RBF: Gaussian(6) = 1
# Outputs: (1*3) + (1*4) = 7
self.assertAlmostEqual(7, y[0], 3)
# Inputs: (2*1) + (2*2) = 6
# RBF: Gaussian(6) = 1
# Outputs: (1*5) + (1*6) = 11
self.assertAlmostEqual(11, y[1], 3)
cls = network.compure_classification(x)
# class 1 is higher than class 0
self.assertEqual(1, cls)
def test_reset_compute(self):
network = RbfNetwork(2, 1, 1)
total = 0
for i in xrange(0, len(network.long_term_memory)):
total += network.long_term_memory[i]
self.assertEquals(0, total)
network.reset()
total = 0
for i in xrange(0, len(network.long_term_memory)):
total += network.long_term_memory[i]
self.assertTrue(total > 1)
def test_reset_compute(self):
network = RbfNetwork(2, 1, 1)
total = 0
for i in xrange(0, len(network.long_term_memory)):
total += network.long_term_memory[i]
self.assertEquals(0, total)
network.reset()
total = 0
for i in xrange(0, len(network.long_term_memory)):
total += network.long_term_memory[i]
self.assertTrue(total > 1)
def test_compute_regression(self):
network = RbfNetwork(2, 1, 1)
network.copy_memory([
2.0, # input 1 to RBF 1
2.0, # input 2 to RBF 1
5.0, # RBF width
2.0, # RBF, center-0
4.0, # RBF, center-1
3.0, # RBF1 to Output 1
4.0 # Bias to Output 1
])
x = [1, 2]
y = network.compute_regression(x)[0]
# Inputs: (2*1) + (2*2) = 6
# RBF: Gaussian(6) = 1
# Outputs: (1*3) + (1*4) = 7
self.assertAlmostEqual(7, y, 3)
def test_compute_regression(self):
network = RbfNetwork(2, 1, 1)
network.copy_memory([
2.0, # input 1 to RBF 1
2.0, # input 2 to RBF 1
5.0, # RBF width
2.0, # RBF, center-0
4.0, # RBF, center-1
3.0, # RBF1 to Output 1
4.0 # Bias to Output 1
])
x = [1, 2]
y = network.compute_regression(x)[0]
# Inputs: (2*1) + (2*2) = 6
# RBF: Gaussian(6) = 1
# Outputs: (1*3) + (1*4) = 7
self.assertAlmostEqual(7, y, 3)
def test_compute_classification(self):
network = RbfNetwork(2, 1, 2)
network.copy_memory([
2.0, # input 1 to RBF 1
2.0, # input 2 to RBF 1
5.0, # RBF width
2.0, # RBF, center-0
4.0, # RBF, center-1
3.0, # RBF1 to Output 1
4.0, # Bias to Output 1
5.0, # RBF1 to Output 2
6.0
]) # Bias to Output 2
x = [1, 2]
y = network.compute_regression(x)
# Inputs: (2*1) + (2*2) = 6
# RBF: Gaussian(6) = 1
# Outputs: (1*3) + (1*4) = 7
self.assertAlmostEqual(7, y[0], 3)
# Inputs: (2*1) + (2*2) = 6
# RBF: Gaussian(6) = 1
# Outputs: (1*5) + (1*6) = 11
self.assertAlmostEqual(11, y[1], 3)
cls = network.compure_classification(x)
# class 1 is higher than class 0
self.assertEqual(1, cls)
inv_classes = {v: k for k, v in classes.items()}
# Normalize iris species using one-of-n.
# We could have used equilateral as well. For an example of equilateral, see the example_nm_iris example.
norm.norm_col_one_of_n(iris_work, 4, classes, 0, 1)
# Prepare training data. Separate into input and ideal.
training = np.array(iris_work)
training_input = training[:, 0:4]
training_ideal = training[:, 4:7]
# Create an RBF network. There are four inputs and two outputs.
# There are also five RBF functions used internally.
# You can experiment with different numbers of internal RBF functions.
# However, the input and output must match the data set.
network = RbfNetwork(4, 4, 3)
network.reset()
def score_funct(x):
"""
The score function for Iris anneal.
@param x:
@return:
"""
global best_score
global input_data
global output_data
# Update the network's long term memory to the vector we need to score.
network.copy_memory(x)
# Loop over the training set and calculate the output for each.
# Normalize iris species using one-of-n.
# We could have used equilateral as well. For an example of equilateral, see the example_nm_iris example.
norm.norm_col_one_of_n(iris_work, 4, classes, 0, 1)
# Prepare training data. Separate into input and ideal.
training = np.array(iris_work)
training_input = training[:, 0:4]
training_ideal = training[:, 4:7]
# Create an RBF network. There are four inputs and two outputs.
# There are also five RBF functions used internally.
# You can experiment with different numbers of internal RBF functions.
# However, the input and output must match the data set.
network = RbfNetwork(4, 4, 3)
network.reset()
def score_funct(x):
"""
The score function for Iris anneal.
@param x:
@return:
"""
global best_score
global input_data
global output_data
# Update the network's long term memory to the vector we need to score.
network.copy_memory(x)
# Loop over the training set and calculate the output for each.
actual_output = []
# The input for the XOR operator.
training_input = np.array([
[0.0, 0.0],
[1.0, 0.0],
[0.0, 1.0],
[1.0, 1.0]])
# The ideal output for the XOR operator. Each row corresponds to a row in the training_input.
training_ideal = np.array([
[0.0],
[1.0],
[1.0],
[0.0],
])
network = RbfNetwork(2, 5, 1)
network.reset()
def score_funct(x):
"""
The score function. Calculate the MSE error between the actual network output and the ideal values for the XOR.
@param x: The long term memory that we are to score.
@return: The MSE error.
"""
global input_data
global output_data
network.copy_memory(x)
actual_output = []
for input_data in training_input:
output_data = network.compute_regression(input_data)
from rbf_network import RbfNetwork
from error import ErrorCalculation
from train import TrainHillClimb
# The input for the XOR operator.
training_input = np.array([[0.0, 0.0], [1.0, 0.0], [0.0, 1.0], [1.0, 1.0]])
# The ideal output for the XOR operator. Each row corresponds to a row in the training_input.
training_ideal = np.array([
[0.0],
[1.0],
[1.0],
[0.0],
])
network = RbfNetwork(2, 5, 1)
network.reset()
def score_funct(x):
"""
The score function. Calculate the MSE error between the actual network output and the ideal values for the XOR.
@param x: The long term memory that we are to score.
@return: The MSE error.
"""
global input_data
global output_data
network.copy_memory(x)
actual_output = []
for input_data in training_input:
output_data = network.compute_regression(input_data)
def test_basics(self):
network = RbfNetwork(2, 1, 1)
# should be 7, (2*1) + (1+(1 bias))*1 + 3 RBF params
# 2 + 2 + 3 = 7
self.assertEquals(7, len(network.long_term_memory))