def test_orgate(self):
p = Perceptron([0.5, 0.5], -0.5)
self.assertEqual(p.activate([0, 0]), 0)
self.assertEqual(p.activate([1, 0]), 1)
self.assertEqual(p.activate([0, 1]), 1)
self.assertEqual(p.activate([1, 1]), 1)
def test_activate(self):
p = Perceptron(2)
p.theta = 1.5
p.w = [1, 1]
self.assertEqual(p.activate([0, 0]), 0)
self.assertEqual(p.activate([0, 1]), 0)
self.assertEqual(p.activate([1, 0]), 0)
self.assertEqual(p.activate([1, 1]), 1)
self.assertEqual(p.activate([2, 0]), 1)
self.assertEqual(p.activate([0, 2]), 1)
def test_train(self):
p = Perceptron(2)
for i in range(100):
p.train([1, 0], 0, 0.05)
p.train([0, 1], 1, 0.05)
self.assertAlmostEqual(p.activate([1, 0]), 0)
self.assertAlmostEqual(p.activate([0, 1]), 1)
def test_train(self):
p = Perceptron(2)
for i in range(100):
p.train([1, 0], 0, 0.05)
p.train([0, 1], 1, 0.05)
self.assertAlmostEqual(p.activate([1, 0]), 0)
self.assertAlmostEqual(p.activate([0, 1]), 1)
def test_partypoort(self):
p = Perceptron([0.6, 0.3, 0.2], -0.4)
self.assertEqual(p.activate([0, 0, 0]), 0)
self.assertEqual(p.activate([1, 0, 0]), 1)
self.assertEqual(p.activate([0, 1, 0]), 0)
self.assertEqual(p.activate([1, 1, 0]), 1)
self.assertEqual(p.activate([0, 0, 1]), 0)
self.assertEqual(p.activate([1, 0, 1]), 1)
self.assertEqual(p.activate([0, 1, 1]), 1)
self.assertEqual(p.activate([1, 1, 1]), 1)
def test_norgate(self):
p = Perceptron([-1, -1, -1], 0)
self.assertEqual(p.activate([0, 0, 0]), 1)
self.assertEqual(p.activate([1, 0, 0]), 0)
self.assertEqual(p.activate([0, 1, 0]), 0)
self.assertEqual(p.activate([1, 1, 0]), 0)
self.assertEqual(p.activate([0, 0, 1]), 0)
self.assertEqual(p.activate([1, 0, 1]), 0)
self.assertEqual(p.activate([0, 1, 1]), 0)
self.assertEqual(p.activate([1, 1, 1]), 0)
def test_activate(self):
p = Perceptron(2)
p.w = [1, 1]
p.theta = 1.5
self.assertEqual(p.activate([0, 0]), 0)
self.assertEqual(p.activate([1, 1]), 1)
self.assertEqual(p.activate([0, 1]), 0)
self.assertEqual(p.activate([1, 0]), 0)
self.assertEqual(p.activate([2, 0]), 1)
self.assertEqual(p.activate([0, 2]), 1)
p.w = [0.1, -0.2]
p.theta = -0.1
self.assertEqual(p.activate([0, 0]), 1)
self.assertEqual(p.activate([0, 2.5]), 0)
self.assertEqual(p.activate([0.3, 0]), 1)
self.assertEqual(p.activate([1.5, 0.5]), 1)
self.assertEqual(p.activate([0.1, 1.5]), 0)
p = Perceptron(5)
p.w = [1, 2, 3, 4, 5]
p.theta = 7
self.assertEqual(p.activate([1, 1, 1, 0, 0]), 0)
self.assertEqual(p.activate([0, 0, 0, 1, 1]), 1)
def test_activate_on_perceptron_with_weights_equal_to_zero_returns_zero(self):
perceptron = Perceptron(bias=0.0, weight_1=0.0, weight_2=0.0)
activation = perceptron.activate(input_1=0.5, input_2=0.5)
self.assertEqual(0.0, activation)
def test_activate_works(self):
perceptron = Perceptron(bias=0.5, weight_1=1.0, weight_2=0.8)
activation = perceptron.activate(input_1=0.4, input_2=0.6)
self.assertEqual(1.38, activation)
def test():
p = Perceptron([0.1, 0.5, 0.1], 2)
output = p.activate([0.0, 0.2, 0.1])
print(output)
def test_invert(self):
p = Perceptron([-1], 0)
self.assertEqual(p.activate([1]), 0)
p = Perceptron([-1], 0)
self.assertEqual(p.activate([0]), 1)
from perceptron import Perceptron
inputs = [-1,-1,1]
weightNum = len(inputs)
weights = 0
learningRate = 0.01
desired = 1
iterate = True
iterCount = 0
ptron = Perceptron()
weights = ptron.setWeights(weightNum)
while iterate:
print "-"*70
print "Initial Weights: ", weights
sum = ptron.activate(weights, inputs)
print "Sum: ", sum
returned = ptron.updateWeights(learningRate, desired, sum)
print "New Weights: ", returned['newWeights']
if not returned['iterate']:
print "Iterations: ", iterCount
iterate = False
print "-"*70
else:
weights = returned['newWeights']
iterCount = iterCount + 1
class PerceptronTest(unittest.TestCase):
def setUp(self):
self.perceptron = Perceptron()
def test_activate_func_returns_positive_one_if_input_is_greater_than_zero(
self
):
actual = self.perceptron.activate(5)
expected = 1
self.assertEqual(actual, expected)
def test_activate_func_returns_negative_one_if_input_is_less_than_zero(
self
):
actual = self.perceptron.activate(-10)
expected = -1
self.assertEqual(actual, expected)
def test_activate_func_returns_negative_one_if_input_is__zero(self):
actual = self.perceptron.activate(0)
expected = -1
self.assertEqual(actual, expected)
def test_get_weights_returns_weights_stored_in_perceptron(self):
expected = [0.5, 0.5, 0.5]
self.perceptron.weights = expected
actual = self.perceptron.get_weights()
self.assertEqual(actual, expected)
@patch('perceptron.random.uniform')
def test_give_weights_random_value_when_create_perceptron(
self,
mock_uniform
):
expected = [0.1, 0.2, 0.3]
mock_uniform.side_effect = expected
perceptron = Perceptron()
actual = perceptron.get_weights()
self.assertEqual(actual, expected)
self.assertEqual(mock_uniform.call_count, 3)
expected_calls = [
call(-1, 1),
call(-1, 1),
call(-1, 1)
]
self.assertEqual(mock_uniform.mock_calls, expected_calls)
@patch('perceptron.random.uniform')
def test_feedforward_returns_positive_one_if_sum_is_greater_than_zero(
self,
mock_uniform
):
expected = [0.1, 0.2, 0.3]
mock_uniform.side_effect = expected
inputs = [1, 1, 1]
perceptron = Perceptron()
actual = perceptron.feedforward(inputs)
expected = 1
self.assertEqual(actual, expected)
@patch('perceptron.random.uniform')
def test_feedforward_returns_negative_one_if_sum_is_less_than_zero(
self,
mock_uniform
):
expected = [0.1, 0.2, 0.3]
mock_uniform.side_effect = expected
inputs = [1, -3, 1]
perceptron = Perceptron()
actual = perceptron.feedforward(inputs)
expected = -1
self.assertEqual(actual, expected)
@patch('perceptron.random.uniform')
def test_train_returns_new_adjusted_weights_in_positive_direction(
self,
mock_uniform
):
expected = [0.1, 0.2, 0.3]
mock_uniform.side_effect = expected
inputs = [1, -3, 1]
desired = 1
perceptron = Perceptron()
actual = perceptron.train(inputs, desired)
expected = [0.12000000000000001, 0.14, 0.32]
self.assertEqual(actual, expected)
@patch('perceptron.random.uniform')
def test_train_returns_new_adjusted_weights_in_negative_direction(
self,
mock_uniform
):
expected = [0.2, 0.1, 0.3]
mock_uniform.side_effect = expected
inputs = [1, -3, 1]
desired = -1
perceptron = Perceptron()
actual = perceptron.train(inputs, desired)
expected = [0.18000000000000002, 0.16, 0.27999999999999997]
self.assertEqual(actual, expected)
