def test_fail_on_nan_data(bad_data_file, input_data_file, output_data_file):
with pytest.raises(ValueError):
ModelFromData(bad_data_file, output_data_file, 1.)
with pytest.raises(ValueError):
ModelFromData(input_data_file, bad_data_file, 1.)
def test_fails_on_duplicate_input_data(input_data_file_with_duplicates,
output_data_file):
with pytest.raises(ValueError):
data_model = ModelFromData(input_data_file_with_duplicates,
output_data_file, 1.)
data_model.evaluate([1, 2, 3, 4, 5])
def test_fail_on_nan_data(bad_data_file, input_data_file, output_data_file):
"""
Ensure an exception is raised if data with any NaNs is provided.
"""
with pytest.raises(ValueError):
ModelFromData(bad_data_file, output_data_file, 1.)
with pytest.raises(ValueError):
ModelFromData(input_data_file, bad_data_file, 1.)
def test_fails_on_unmatched_input(input_data_file, output_data_file):
"""
Ensure an exception is raised if an input supplied to evaluate does not
have a match in the input data.
"""
data_model = ModelFromData(input_data_file, output_data_file, 1.)
with pytest.raises(ValueError):
data_model.evaluate(-1)
def test_single_row_2d_input_data(input_data_file_2d, output_data_file_2d):
"""
Test ability to evaluate one row of input data passed in as 2d array.
"""
# Initialize model from spring-mass example data files:
data_model = ModelFromData(input_data_file_2d, output_data_file_2d, 1.)
# Should be able to do this without an exception.
one_row_2d_input = np.array([data_model._inputs[0]])
data_model.evaluate(one_row_2d_input)
def test_fails_on_duplicate_input_data(input_data_file_with_duplicates,
output_data_file):
"""
Ensure an exception is raised if there are duplicate inputs found in the
input data.
"""
with pytest.raises(ValueError):
data_model = ModelFromData(input_data_file_with_duplicates,
output_data_file, 1.)
data_model.evaluate([1, 2, 3, 4, 5])
def test_evaluate_2d_data(input_data_file_2d, output_data_file_2d, index):
# Initialize model from spring-mass example data files:
data_model = ModelFromData(input_data_file_2d, output_data_file_2d, 1.)
input_data = np.genfromtxt(input_data_file_2d)
output_data = np.genfromtxt(output_data_file_2d)
# Model expects arrays as inputs/outputs
model_output = data_model.evaluate(input_data[index])
true_output = output_data[index]
assert np.all(np.isclose(model_output, true_output))
def test_evaluate_with_cost_delay(cost, input_data_file, output_data_file):
model = ModelFromData(input_data_file, output_data_file, cost=cost,
wait_cost_duration=True)
sample = model._inputs[0]
start_time = timeit.default_timer()
model.evaluate(sample)
evaluation_time = timeit.default_timer() - start_time
# Ensure evaluation time was close to specified cost.
assert np.abs(evaluation_time - cost) < .01
def test_skip_rows(input_data_file_2d, output_data_file_2d, rows_to_skip):
normal_model = ModelFromData(input_data_file_2d, output_data_file_2d, 1.)
normal_row_count = normal_model._inputs.shape[0]
skipped_row_model = ModelFromData(input_data_file_2d,
output_data_file_2d,
skip_header=rows_to_skip,
cost=1.)
skipped_row_count = skipped_row_model._inputs.shape[0]
assert normal_row_count - rows_to_skip == skipped_row_count
def test_evaluate_1d_data(input_data_file, output_data_file, index):
"""
Test ability to evaluate one dimensional data.
"""
# Initialize model from spring-mass example data files:
data_model = ModelFromData(input_data_file, output_data_file, 1.)
input_data = np.genfromtxt(input_data_file)
output_data = np.genfromtxt(output_data_file)
# Model expects arrays as inputs/outputs.
model_output = data_model.evaluate(input_data[index])
true_output = output_data[index]
assert np.all(np.isclose(model_output, true_output))
def models_from_2d_data():
"""
Creates a list of three ModelFromData objects with a small amount of
two dimensional data.
"""
input_filepath = os.path.join(data_path, "2D_test_data.csv")
output1_filepath = os.path.join(data_path, "2D_test_data_output.csv")
output2_filepath = os.path.join(data_path, "2D_test_data_output.csv")
output3_filepath = os.path.join(data_path, "2D_test_data_output.csv")
model1 = ModelFromData(input_filepath, output1_filepath, 1.)
model2 = ModelFromData(input_filepath, output2_filepath, 4.)
model3 = ModelFromData(input_filepath, output3_filepath, 16.)
return [model1, model2, model3]
def models_from_data():
"""
Creates a list of three ModelFromData objects of increasing fidelity.
:return:
"""
input_filepath = os.path.join(data_path, "spring_mass_1D_inputs.txt")
output1_filepath = os.path.join(data_path, "spring_mass_1D_outputs_1.0.txt")
output2_filepath = os.path.join(data_path, "spring_mass_1D_outputs_0.1.txt")
output3_filepath = os.path.join(data_path,
"spring_mass_1D_outputs_0.01.txt")
model1 = ModelFromData(input_filepath, output1_filepath, 1.)
model2 = ModelFromData(input_filepath, output2_filepath, 4.)
model3 = ModelFromData(input_filepath, output3_filepath, 16.)
return [model1, model2, model3]
def test_fail_if_model_outputs_do_not_match_shapes(filename_2d_5_column_data,
filename_2d_3_column_data):
"""
Ensures simulator throws an exception if inputs and outputs with differing
numbers of samples are provided.
"""
model1 = ModelFromData(filename_2d_5_column_data,
filename_2d_5_column_data, 1.)
model2 = ModelFromData(filename_2d_5_column_data,
filename_2d_3_column_data, 4.)
data_input = InputFromData(filename_2d_5_column_data)
with pytest.raises(ValueError):
MLMCSimulator(models=[model1, model2], data=data_input)
def test_input_output_with_differing_column_count(filename_2d_5_column_data,
filename_2d_3_column_data):
"""
Ensures that simulator handles input and output data with differing numbers
of columns.
"""
model1 = ModelFromData(filename_2d_5_column_data,
filename_2d_3_column_data, 1.)
model2 = ModelFromData(filename_2d_5_column_data,
filename_2d_3_column_data, 4.)
data_input = InputFromData(filename_2d_5_column_data)
sim = MLMCSimulator(models=[model1, model2], data=data_input)
sim.simulate(100., 10)
def test_evaluate_with_cost_delay(cost, input_data_file, output_data_file):
"""
Test ability to wait for the duration of the cost specified for the model
when evaluating a sample.
"""
model = ModelFromData(input_data_file,
output_data_file,
cost=cost,
wait_cost_duration=True)
sample = model._inputs[0]
start_time = timeit.default_timer()
model.evaluate(sample)
evaluation_time = timeit.default_timer() - start_time
# Ensure evaluation time was close to specified cost.
assert np.abs(evaluation_time - cost) < .01
def test_evaluate_fails_on_invalid_input(input_data_file, output_data_file):
data_model = ModelFromData(input_data_file, output_data_file, 1.)
bogus_input = "five"
with pytest.raises(TypeError):
data_model.evaluate(bogus_input)
bogus_input = [[1, 2], [3, 4]]
with pytest.raises(ValueError):
data_model.evaluate(bogus_input)
def test_evaluate_fails_on_invalid_input(input_data_file, output_data_file):
"""
Ensure that an invalid input triggers an exception in evaluate.
"""
data_model = ModelFromData(input_data_file, output_data_file, 1.)
bogus_input = "five"
with pytest.raises(TypeError):
data_model.evaluate(bogus_input)
bogus_input = [[1, 2], [3, 4]]
with pytest.raises(ValueError):
data_model.evaluate(bogus_input)
def test_init_fails_on_invalid_cost(input_data_file, output_data_file, cost):
with pytest.raises(Exception):
ModelFromData(input_data_file, output_data_file, cost)
from MLMCPy.mlmc import MLMCSimulator
from MLMCPy.input import InputFromData
from MLMCPy.model import ModelFromData
my_path = os.path.dirname(os.path.abspath(__file__))
data_path = my_path + "/../../tests/testing_data"
data_input = InputFromData(os.path.join(data_path,
"spring_mass_1D_inputs.txt"),
shuffle_data=False)
input_filepath = os.path.join(data_path, "spring_mass_1D_inputs.txt")
output1_filepath = os.path.join(data_path, "spring_mass_1D_outputs_1.0.txt")
output2_filepath = os.path.join(data_path, "spring_mass_1D_outputs_0.1.txt")
output3_filepath = os.path.join(data_path, "spring_mass_1D_outputs_0.01.txt")
model1 = ModelFromData(input_filepath, output1_filepath, 1.)
model2 = ModelFromData(input_filepath, output2_filepath, 4.)
model3 = ModelFromData(input_filepath, output3_filepath, 16.)
models_from_data = [model1, model2, model3]
np.random.seed(1)
initial_sample_size = 200
epsilon = 1.
# Get output data for each layer.
level_0_data = np.zeros(initial_sample_size)
level_1_data = np.zeros(initial_sample_size)
input_samples = data_input.draw_samples(initial_sample_size)
from MLMCPy.input import InputFromData
from MLMCPy.mlmc import MLMCSimulator
from MLMCPy.model import ModelFromData
# Define I/O files
inputfile = "data/spring_mass_1D_inputs.txt"
outputfile_level1 = "data/spring_mass_1D_outputs_1.0.txt"
outputfile_level2 = "data/spring_mass_1D_outputs_0.1.txt"
outputfile_level3 = "data/spring_mass_1D_outputs_0.01.txt"
# Initialize random input & model objects
data_input = InputFromData(inputfile)
model_level1 = ModelFromData(inputfile, outputfile_level1, cost=1.0)
model_level2 = ModelFromData(inputfile, outputfile_level2, cost=10.0)
model_level3 = ModelFromData(inputfile, outputfile_level3, cost=100.0)
models = [model_level1, model_level2, model_level3]
mlmc_simulator = MLMCSimulator(data_input, models)
[estimates, sample_sizes, variances] = \
mlmc_simulator.simulate(epsilon=1e-1,
initial_sample_sizes=100)
print 'Estimate: %s' % estimates
print 'Sample sizes used: %s' % sample_sizes
print 'Variance: %s' % variances
def test_init_fails_on_invalid_cost(input_data_file, output_data_file, cost):
"""
Ensure that an invalid cost parameter triggers an exception in init.
"""
with pytest.raises(Exception):
ModelFromData(input_data_file, output_data_file, cost)
def test_init_fails_on_incompatible_data(input_data_file, output_data_file_2d):
"""
Ensure an exception is raised if input and output data are incompatible.
"""
with pytest.raises(ValueError):
ModelFromData(input_data_file, output_data_file_2d, 1.)
def test_init_fails_on_incompatible_data(input_data_file, output_data_file_2d):
with pytest.raises(ValueError):
ModelFromData(input_data_file, output_data_file_2d, 1.)