def test_four(in_sample, out_sample):
training_set = random_set(in_sample, moved_circle)
noisy_indices = np.random.choice(in_sample, size=round(0.1 * in_sample), replace=False)
training_set.y[noisy_indices] *= -1
weight = linear_percepton(training_set.z, training_set.y)
in_error_no_transform = weight_error(weight, training_set.z, training_set.y)
training_set.z = second_order(training_set.z)
weight = linear_percepton(training_set.z, training_set.y)
in_error_transform = weight_error(weight, training_set.z, training_set.y)
testing_set = random_set(out_sample, moved_circle, second_order)
noisy_indices = np.random.choice(out_sample, size=round(0.1 * out_sample), replace=False)
testing_set.y[noisy_indices] *= -1
out_error_transform = weight_error(weight, testing_set.z, testing_set.y)
return in_error_no_transform, weight, out_error_transform
def test1(training_data, testing_data):
training_set = DataML(training_data, transform)
weight = linear_percepton(training_set.z, training_set.y)
testing_set = DataML(testing_data, transform)
in_error, out_error = [ weight_error(weight, data_set.z, data_set.y)
for data_set in [training_set, testing_set] ]
return in_error, out_error
def get_line_to_parameters(trials):
"""line_to : line through origin"""
gradients = np.array([
linear_percepton(training_set.x, training_set.y)
for training_set in trials
])
return gradients
def test_two(in_sample, out_sample):
target_function = random_target_function()
training_set = random_set(in_sample, target_function)
weight = linear_percepton(training_set.z, training_set.y)
in_error = weight_error(weight, training_set.z, training_set.y)
testing_set = random_set(out_sample, target_function)
out_error = weight_error(weight, testing_set.z, testing_set.y)
return in_error, out_error
def test_two(in_sample, out_sample):
target_function = random_target_function()
training_set = random_set(in_sample, target_function)
weight = linear_percepton(training_set.z, training_set.y)
in_error = weight_error(weight, training_set.z, training_set.y)
testing_set = random_set(out_sample, target_function)
out_error = weight_error(weight, testing_set.z, testing_set.y)
return in_error, out_error
def gen_models(training_set):
k_values = range(
3, 8
) # k values from question. 8 instead of 7 because range is not inclsuive
weights = [
linear_percepton(training_set.z[:, :k + 1], training_set.y)
for k in k_values
] # k+1 as bound is not inclusive
return weights
def test1(training_data, testing_data):
training_set = DataML(training_data, transform)
weight = linear_percepton(training_set.z, training_set.y)
testing_set = DataML(testing_data, transform)
in_error, out_error = [
weight_error(weight, data_set.z, data_set.y)
for data_set in [training_set, testing_set]
]
return in_error, out_error
def test_four(in_sample, out_sample):
training_set = random_set(in_sample, moved_circle)
noisy_indices = np.random.choice(in_sample,
size=round(0.1 * in_sample),
replace=False)
training_set.y[noisy_indices] *= -1
weight = linear_percepton(training_set.z, training_set.y)
in_error_no_transform = weight_error(weight, training_set.z,
training_set.y)
training_set.z = second_order(training_set.z)
weight = linear_percepton(training_set.z, training_set.y)
in_error_transform = weight_error(weight, training_set.z, training_set.y)
testing_set = random_set(out_sample, moved_circle, second_order)
noisy_indices = np.random.choice(out_sample,
size=round(0.1 * out_sample),
replace=False)
testing_set.y[noisy_indices] *= -1
out_error_transform = weight_error(weight, testing_set.z, testing_set.y)
return in_error_no_transform, weight, out_error_transform
def get_line_parameters(trials):
new_trials = [
DataML((np.insert(training_set.z, 0, 1, axis=1), training_set.y))
for training_set in trials
]
weights = np.array([
linear_percepton(training_set.z, training_set.y)
for training_set in new_trials
])
return weights
def generate_quadratic_to_parameters(trials):
"""line_to : line through origin
function of form ax^2
"""
def transform(x):
return x ** 2
new_trials = [ DataML((training_set.z, training_set.y), transform)
for training_set in trials ]
weights = np.array([ linear_percepton(training_set.x, training_set.y) for training_set in new_trials ])
return weights
def generate_quadratic_to_parameters(trials):
"""line_to : line through origin
function of form ax^2
"""
def transform(x):
return x**2
new_trials = [
DataML((training_set.z, training_set.y), transform)
for training_set in trials
]
weights = np.array([
linear_percepton(training_set.x, training_set.y)
for training_set in new_trials
])
return weights
def generate_quadratic_parameters(trials):
"""ax^2 + b"""
def transform(x):
"""
transform
x1 ---> 1 x1**2
"""
ones = np.ones(len(x))
x1 = x[:, 0]
x1_sqr = x1 ** 2
return np.stack([ones, x1_sqr], axis=1)
new_trials = [
DataML((training_set.z, training_set.y), transform)
for training_set in trials ]
weights = [ linear_percepton(training_set.z, training_set.y) for training_set in new_trials ]
return np.array(weights)
def generate_quadratic_parameters(trials):
"""ax^2 + b"""
def transform(x):
"""
transform
x1 ---> 1 x1**2
"""
ones = np.ones(len(x))
x1 = x[:, 0]
x1_sqr = x1**2
return np.stack([ones, x1_sqr], axis=1)
new_trials = [
DataML((training_set.z, training_set.y), transform)
for training_set in trials
]
weights = [
linear_percepton(training_set.z, training_set.y)
for training_set in new_trials
]
return np.array(weights)
def gen_models(training_set):
k_values = range(3,8) # k values from question. 8 instead of 7 because range is not inclsuive
weights = [ linear_percepton(training_set.z[:,:k + 1], training_set.y)
for k in k_values ] # k+1 as bound is not inclusive
return weights
def get_line_parameters(trials):
new_trials = [
DataML((np.insert(training_set.z, 0, 1, axis=1), training_set.y))
for training_set in trials ]
weights = np.array([ linear_percepton(training_set.z, training_set.y) for training_set in new_trials ])
return weights
def get_line_to_parameters(trials):
"""line_to : line through origin"""
gradients = np.array([ linear_percepton(training_set.x, training_set.y) for training_set in trials ])
return gradients