def test_model_stacking_transform():
model_stacking = Pipeline([
ModelStacking(
[
SKLearnWrapper(
GradientBoostingRegressor(),
HyperparameterSpace({
"n_estimators": RandInt(50, 600),
"max_depth": RandInt(1, 10),
"learning_rate": LogUniform(0.07, 0.7)
})),
SKLearnWrapper(
KMeans(),
HyperparameterSpace({"n_clusters": RandInt(5, 10)})),
],
joiner=NumpyTranspose(),
judge=SKLearnWrapper(
Ridge(),
HyperparameterSpace({
"alpha": LogUniform(0.7, 1.4),
"fit_intercept": Boolean()
})),
)
])
expected_outputs_shape = (379, 1)
data_inputs_shape = (379, 13)
data_inputs = _create_data(data_inputs_shape)
expected_outputs = _create_data(expected_outputs_shape)
model_stacking = model_stacking.fit(data_inputs, expected_outputs)
outputs = model_stacking.transform(data_inputs)
assert outputs.shape == expected_outputs_shape
def test_automl_should_shallow_copy_data_before_each_epoch():
# see issue #332 https://github.com/Neuraxio/Neuraxle/issues/332
data_inputs = np.random.randint(0, 100, (100, 3))
expected_outputs = np.random.randint(0, 3, 100)
from sklearn.preprocessing import StandardScaler
p = Pipeline([
SKLearnWrapper(StandardScaler()),
SKLearnWrapper(LinearSVC(),
HyperparameterSpace({'C': RandInt(0, 10000)})),
])
auto_ml = AutoML(p,
validation_splitter=ValidationSplitter(0.20),
refit_trial=True,
n_trials=10,
epochs=10,
cache_folder_when_no_handle='cache',
scoring_callback=ScoringCallback(
mean_squared_error, higher_score_is_better=False),
callbacks=[
MetricCallback('mse',
metric_function=mean_squared_error,
higher_score_is_better=False)
],
hyperparams_repository=InMemoryHyperparamsRepository(
cache_folder='cache'),
continue_loop_on_error=False)
random_search = auto_ml.fit(data_inputs, expected_outputs)
best_model = random_search.get_best_model()
assert isinstance(best_model, Pipeline)
def test_automl_sequential_wrapper_with_validation_split_wrapper(tmpdir):
# Setting seed for reproducibility
np.random.seed(75)
# Given
data_inputs = np.array(range(100))
expected_outputs = np.array(range(100, 200))
hyperparameter_space = HyperparameterSpace({
'multiplication_1__multiply_by':
RandInt(1, 3),
'multiplication_2__multiply_by':
RandInt(1, 3),
'multiplication_3__multiply_by':
RandInt(1, 3),
})
pipeline = Pipeline(
[('multiplication_1', MultiplyByN()),
('multiplication_2', MultiplyByN()),
('multiplication_3', MultiplyByN())],
cache_folder=tmpdir).set_hyperparams_space(hyperparameter_space)
random_search = RandomSearch(
ValidationSplitWrapper(pipeline,
test_size=0.2,
scoring_function=mean_squared_error,
run_validation_split_in_test_mode=False),
hyperparams_repository=HyperparamsJSONRepository(tmpdir),
higher_score_is_better=False,
n_iter=100)
# When
mse_before = ((data_inputs - expected_outputs)**2).mean()
random_search: AutoMLSequentialWrapper = random_search.fit(
data_inputs, expected_outputs)
best_model: Pipeline = random_search.get_best_model()
predicted_outputs = best_model.transform(data_inputs)
# Then
actual_mse = ((predicted_outputs - expected_outputs)**2).mean()
assert actual_mse < mse_before
def main():
p = Pipeline([('step1', MultiplyByN()), ('step2', MultiplyByN()),
Pipeline([Identity(),
Identity(),
PCA(n_components=4)])])
p.set_hyperparams_space({
'step1__multiply_by': RandInt(42, 50),
'step2__multiply_by': RandInt(-10, 0),
'Pipeline__PCA__n_components': RandInt(2, 3)
})
samples = p.get_hyperparams_space().rvs()
p.set_hyperparams(samples)
samples = p.get_hyperparams()
assert 42 <= samples['step1__multiply_by'] <= 50
assert -10 <= samples['step2__multiply_by'] <= 0
assert samples['Pipeline__PCA__n_components'] in [2, 3]
assert p['Pipeline']['PCA'].get_wrapped_sklearn_predictor(
).n_components in [2, 3]
def test_hyperparam_space():
p = Pipeline([
AddFeatures([
SomeStep(hyperparams_space=HyperparameterSpace({"n_components": RandInt(1, 5)})),
SomeStep(hyperparams_space=HyperparameterSpace({"n_components": RandInt(1, 5)}))
]),
ModelStacking([
SomeStep(hyperparams_space=HyperparameterSpace({"n_estimators": RandInt(1, 1000)})),
SomeStep(hyperparams_space=HyperparameterSpace({"n_estimators": RandInt(1, 1000)})),
SomeStep(hyperparams_space=HyperparameterSpace({"max_depth": RandInt(1, 100)})),
SomeStep(hyperparams_space=HyperparameterSpace({"max_depth": RandInt(1, 100)}))
],
joiner=NumpyTranspose(),
judge=SomeStep(hyperparams_space=HyperparameterSpace({"alpha": LogUniform(0.1, 10.0)}))
)
])
rvsed = p.get_hyperparams_space()
p.set_hyperparams(rvsed)
hyperparams = p.get_hyperparams()
assert "AddFeatures" in hyperparams.keys()
assert "SomeStep" in hyperparams["AddFeatures"]
assert "n_components" in hyperparams["AddFeatures"]["SomeStep"]
assert "SomeStep1" in hyperparams["AddFeatures"]
assert "n_components" in hyperparams["AddFeatures"]["SomeStep1"]
assert "SomeStep" in hyperparams["ModelStacking"]
assert "n_estimators" in hyperparams["ModelStacking"]["SomeStep"]
assert "SomeStep1" in hyperparams["ModelStacking"]
assert "max_depth" in hyperparams["ModelStacking"]["SomeStep2"]
def test_automl_sklearn_model_with_base_estimator(tmpdir):
grad_boost = GradientBoostingRegressor()
bagged_regressor = BaggingRegressor(grad_boost, random_state=5, n_jobs=-1)
wrapped_bagged_regressor = SKLearnWrapper(
bagged_regressor,
HyperparameterSpace({
"n_estimators": RandInt(10, 100),
"max_features": Uniform(0.6, 1.0)
}),
# return_all_sklearn_default_params_on_get=True
)
_test_within_auto_ml_loop(tmpdir, wrapped_bagged_regressor)
def test_meta_step_mixin_update_hyperparams_space_should_update_wrapped_step_hyperparams():
p = SomeMetaStepMixin(SomeStep())
p.set_hyperparams_space(HyperparameterSpace({
META_STEP_HP: RAND_INT_META_STEP,
SOME_STEP_HP: RAND_INT_SOME_STEP
}))
updated_some_step_hp_space = RandInt(0, 100)
p.update_hyperparams_space(HyperparameterSpace({
SOME_STEP_HP: updated_some_step_hp_space
}))
assert p.hyperparams_space[META_STEP_HP] == RAND_INT_META_STEP
assert p.wrapped.get_hyperparams_space()['somestep_hyperparam'] == updated_some_step_hp_space
def test_pipeline_should_get_hyperparams_space():
p = Pipeline([
SomeStep().set_name('step_1'),
SomeStep().set_name('step_2')
])
p.set_hyperparams_space({
'hp': RandInt(1, 2),
'step_1__hp': RandInt(2, 3),
'step_2__hp': RandInt(3, 4)
})
hyperparams_space = p.get_hyperparams_space()
assert isinstance(hyperparams_space, HyperparameterSpace)
assert hyperparams_space['hp'].min_included == 1
assert hyperparams_space['hp'].max_included == 2
assert hyperparams_space['step_1__hp'].min_included == 2
assert hyperparams_space['step_1__hp'].max_included == 3
assert hyperparams_space['step_2__hp'].min_included == 3
assert hyperparams_space['step_2__hp'].max_included == 4
def test_step_cloner_update_hyperparams_space_should_update_wrapped_step_hyperparams():
p = StepClonerForEachDataInput(SomeStep())
p.set_hyperparams_space(HyperparameterSpace({
META_STEP_HP: RAND_INT_META_STEP,
SOME_STEP_HP: RAND_INT_SOME_STEP
}))
updated_some_step_hp_space = RandInt(0, 400)
p.update_hyperparams_space(HyperparameterSpace({
SOME_STEP_HP: updated_some_step_hp_space
}))
assert isinstance(p.hyperparams, HyperparameterSamples)
assert p.hyperparams_space[META_STEP_HP] == RAND_INT_META_STEP
assert p.wrapped.get_hyperparams_space()[SOME_STEP_HP_KEY] == updated_some_step_hp_space
def test_automl_sequential_wrapper(tmpdir):
# Setting seed for reproducibility
np.random.seed(68)
# Given
data_inputs = np.array(range(100))
expected_outputs = np.array(range(100, 200))
hyperparameter_space = HyperparameterSpace({
'multiplication_1__multiply_by':
RandInt(1, 3),
'multiplication_2__multiply_by':
RandInt(1, 3),
'multiplication_3__multiply_by':
RandInt(1, 3),
})
pipeline = Pipeline(
[('multiplication_1', MultiplyByN()),
('multiplication_2', MultiplyByN()),
('multiplication_3', MultiplyByN())],
cache_folder=tmpdir).set_hyperparams_space(hyperparameter_space)
auto_ml = RandomSearch(
KFoldCrossValidationWrapper().set_step(pipeline),
hyperparams_repository=HyperparamsJSONRepository(tmpdir),
n_iter=10)
# When
auto_ml: AutoMLSequentialWrapper = auto_ml.fit(data_inputs,
expected_outputs)
best_model: Pipeline = auto_ml.get_best_model()
predicted_outputs = best_model.transform(data_inputs)
# Then
actual_mse = ((predicted_outputs - expected_outputs)**2).mean()
assert actual_mse < 20000
def test_hyperparam_space():
p = Pipeline([
AddFeatures([
SomeStep(hyperparams_space=HyperparameterSpace(
{"n_components": RandInt(1, 5)})),
SomeStep(hyperparams_space=HyperparameterSpace(
{"n_components": RandInt(1, 5)}))
]),
ModelStacking([
SomeStep(hyperparams_space=HyperparameterSpace(
{"n_estimators": RandInt(1, 1000)})),
SomeStep(hyperparams_space=HyperparameterSpace(
{"n_estimators": RandInt(1, 1000)})),
SomeStep(hyperparams_space=HyperparameterSpace(
{"max_depth": RandInt(1, 100)})),
SomeStep(hyperparams_space=HyperparameterSpace(
{"max_depth": RandInt(1, 100)}))
],
joiner=NumpyTranspose(),
judge=SomeStep(hyperparams_space=HyperparameterSpace(
{"alpha": LogUniform(0.1, 10.0)})))
])
rvsed = p.get_hyperparams_space()
p.set_hyperparams(rvsed)
hyperparams = p.get_hyperparams()
flat_hyperparams_keys = hyperparams.to_flat_dict().keys()
assert 'AddFeatures' in hyperparams
assert 'SomeStep' in hyperparams["AddFeatures"]
assert "n_components" in hyperparams["AddFeatures"]["SomeStep"]
assert 'SomeStep1' in hyperparams["AddFeatures"]
assert "n_components" in hyperparams["AddFeatures"]["SomeStep1"]
assert 'ModelStacking' in hyperparams
assert 'SomeStep' in hyperparams["ModelStacking"]
assert 'n_estimators' in hyperparams["ModelStacking"]["SomeStep"]
assert 'SomeStep1' in hyperparams["ModelStacking"]
assert 'n_estimators' in hyperparams["ModelStacking"]["SomeStep1"]
assert 'SomeStep2' in hyperparams["ModelStacking"]
assert 'max_depth' in hyperparams["ModelStacking"]["SomeStep2"]
assert 'SomeStep3' in hyperparams["ModelStacking"]
assert 'max_depth' in hyperparams["ModelStacking"]["SomeStep3"]
assert 'AddFeatures__SomeStep1__n_components' in flat_hyperparams_keys
assert 'AddFeatures__SomeStep__n_components' in flat_hyperparams_keys
assert 'ModelStacking__SomeStep__n_estimators' in flat_hyperparams_keys
assert 'ModelStacking__SomeStep1__n_estimators' in flat_hyperparams_keys
assert 'ModelStacking__SomeStep2__max_depth' in flat_hyperparams_keys
assert 'ModelStacking__SomeStep3__max_depth' in flat_hyperparams_keys
from neuraxle.base import BaseStep, TruncableSteps, NonFittableMixin, MetaStepMixin
from neuraxle.hyperparams.distributions import LogUniform, Quantized, RandInt, Boolean
from neuraxle.hyperparams.space import HyperparameterSpace, HyperparameterSamples
HYPERPARAMETERS_SPACE = HyperparameterSpace({
'learning_rate':
LogUniform(0.0001, 0.1),
'l2_weight_reg':
LogUniform(0.0001, 0.1),
'momentum':
LogUniform(0.01, 1.0),
'hidden_size':
Quantized(LogUniform(16, 512)),
'num_layers':
RandInt(1, 4),
'num_lstm_layers':
RandInt(1, 2),
'use_xavier_init':
Boolean(),
'use_max_pool_else_avg_pool':
Boolean(),
'dropout_drop_proba':
LogUniform(0.3, 0.7)
})
HYPERPARAMETERS = HyperparameterSamples({
'learning_rate': 0.1,
'l2_weight_reg': 0.001,
'hidden_size': 32,
'num_layers': 3,
'num_lstm_layers': 1,
boston = load_boston()
X, y = shuffle(boston.data, boston.target, random_state=13)
X = X.astype(np.float32)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=0.25,
shuffle=False)
# Note that the hyperparameter spaces are defined here during the pipeline definition, but it could be already set
# within the classes ar their definition if using custom classes, or also it could be defined after declaring the
# pipeline using a flat dict or a nested dict.
p = Pipeline([
AddFeatures([
SKLearnWrapper(PCA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})),
SKLearnWrapper(FastICA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})),
]),
ModelStacking(
[
SKLearnWrapper(
GradientBoostingRegressor(),
HyperparameterSpace({
"n_estimators": RandInt(50, 600),
"max_depth": RandInt(1, 10),
"learning_rate": LogUniform(0.07, 0.7)
})),
SKLearnWrapper(
GradientBoostingRegressor(),
HyperparameterSpace({
from neuraxle.base import MetaStepMixin, BaseStep, NonFittableMixin, NonTransformableMixin
from neuraxle.hyperparams.distributions import RandInt, Boolean
from neuraxle.hyperparams.space import HyperparameterSpace, HyperparameterSamples
from neuraxle.steps.loop import StepClonerForEachDataInput
from testing.test_pipeline import SomeStep
SOME_STEP_HP_KEY = 'somestep_hyperparam'
RAND_INT_SOME_STEP = RandInt(-10, 0)
RAND_INT_STEP_CLONER = RandInt(0, 10)
META_STEP_HP = 'metastep_hyperparam'
SOME_STEP_HP = "SomeStep__somestep_hyperparam"
META_STEP_HP_VALUE = 1
SOME_STEP_HP_VALUE = 2
HYPE_SPACE = HyperparameterSpace({"a__test": Boolean()})
HYPE_SAMPLE = HyperparameterSamples({"a__test": True})
class SomeMetaStepMixin(NonTransformableMixin, NonFittableMixin, MetaStepMixin,
BaseStep):
pass
class SomeStepInverseTransform(SomeStep):
def fit_transform(self, data_inputs, expected_outputs=None):
return self, 'fit_transform'
def inverse_transform(self, processed_outputs):
return 'inverse_transform'
def test_deep_learning_pipeline():
# Given
boston = load_boston()
data_inputs, expected_outputs = shuffle(boston.data,
boston.target,
random_state=13)
expected_outputs = expected_outputs.astype(np.float32)
data_inputs = data_inputs.astype(np.float32)
pipeline = Pipeline([
AddFeatures([
SKLearnWrapper(
PCA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})),
SKLearnWrapper(
FastICA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})),
]),
ModelStacking(
[
SKLearnWrapper(
GradientBoostingRegressor(),
HyperparameterSpace({
"n_estimators": RandInt(50, 600),
"max_depth": RandInt(1, 10),
"learning_rate": LogUniform(0.07, 0.7)
})),
SKLearnWrapper(
KMeans(n_clusters=7),
HyperparameterSpace({"n_clusters": RandInt(5, 10)})),
],
joiner=NumpyTranspose(),
judge=SKLearnWrapper(
Ridge(),
HyperparameterSpace({
"alpha": LogUniform(0.7, 1.4),
"fit_intercept": Boolean()
})),
)
])
p = DeepLearningPipeline(
pipeline,
validation_size=VALIDATION_SIZE,
batch_size=BATCH_SIZE,
batch_metrics={'mse': to_numpy_metric_wrapper(mean_squared_error)},
shuffle_in_each_epoch_at_train=True,
n_epochs=N_EPOCHS,
epochs_metrics={'mse': to_numpy_metric_wrapper(mean_squared_error)},
scoring_function=to_numpy_metric_wrapper(mean_squared_error),
)
# When
p, outputs = p.fit_transform(data_inputs, expected_outputs)
# Then
batch_mse_train = p.get_batch_metric_train('mse')
epoch_mse_train = p.get_epoch_metric_train('mse')
batch_mse_validation = p.get_batch_metric_validation('mse')
epoch_mse_validation = p.get_epoch_metric_validation('mse')
assert len(epoch_mse_train) == N_EPOCHS
assert len(epoch_mse_validation) == N_EPOCHS
expected_len_batch_mse_train = math.ceil(
(len(data_inputs) / BATCH_SIZE) * (1 - VALIDATION_SIZE)) * N_EPOCHS
expected_len_batch_mse_validation = math.ceil(
(len(data_inputs) / BATCH_SIZE) * VALIDATION_SIZE) * N_EPOCHS
assert len(batch_mse_train) == expected_len_batch_mse_train
assert len(batch_mse_validation) == expected_len_batch_mse_validation
last_batch_mse_validation = batch_mse_validation[-1]
last_batch_mse_train = batch_mse_train[-1]
last_epoch_mse_train = epoch_mse_train[-1]
last_epoch_mse_validation = epoch_mse_validation[-1]
assert last_batch_mse_train < last_batch_mse_validation
assert last_epoch_mse_train < last_epoch_mse_validation
assert last_batch_mse_train < 1
assert last_epoch_mse_train < 1
def main(tmpdir, sleep_time: float = 0.001, n_iter: int = 10):
DATA_INPUTS = np.array(range(100))
EXPECTED_OUTPUTS = np.array(range(100, 200))
HYPERPARAMETER_SPACE = HyperparameterSpace({
'multiplication_1__multiply_by': RandInt(1, 2),
'multiplication_2__multiply_by': RandInt(1, 2),
'multiplication_3__multiply_by': RandInt(1, 2),
})
print('Classic Pipeline:')
classic_pipeline_folder = os.path.join(str(tmpdir), 'classic')
pipeline = Pipeline([
('multiplication_1', MultiplyByN()),
('sleep_1', ForEachDataInput(Sleep(sleep_time))),
('multiplication_2', MultiplyByN()),
('sleep_2', ForEachDataInput(Sleep(sleep_time))),
('multiplication_3', MultiplyByN()),
], cache_folder=classic_pipeline_folder).set_hyperparams_space(HYPERPARAMETER_SPACE)
time_a = time.time()
auto_ml = AutoML(
pipeline,
refit_trial=True,
n_trials=n_iter,
cache_folder_when_no_handle=classic_pipeline_folder,
validation_splitter=ValidationSplitter(0.2),
hyperparams_optimizer=RandomSearchHyperparameterSelectionStrategy(),
scoring_callback=ScoringCallback(mean_squared_error, higher_score_is_better=False),
callbacks=[
MetricCallback('mse', metric_function=mean_squared_error, higher_score_is_better=False)
],
)
auto_ml = auto_ml.fit(DATA_INPUTS, EXPECTED_OUTPUTS)
outputs = auto_ml.get_best_model().predict(DATA_INPUTS)
time_b = time.time()
actual_score = mean_squared_error(EXPECTED_OUTPUTS, outputs)
print('{0} seconds'.format(time_b - time_a))
print('output: {0}'.format(outputs))
print('smallest mse: {0}'.format(actual_score))
print('best hyperparams: {0}'.format(pipeline.get_hyperparams()))
assert isinstance(actual_score, float)
print('Resumable Pipeline:')
resumable_pipeline_folder = os.path.join(str(tmpdir), 'resumable')
pipeline = ResumablePipeline([
('multiplication_1', MultiplyByN()),
('ForEach(sleep_1)', ForEachDataInput(Sleep(sleep_time))),
('checkpoint1', ExpandDim(DefaultCheckpoint())),
('multiplication_2', MultiplyByN()),
('sleep_2', ForEachDataInput(Sleep(sleep_time))),
('checkpoint2', ExpandDim(DefaultCheckpoint())),
('multiplication_3', MultiplyByN())
], cache_folder=resumable_pipeline_folder).set_hyperparams_space(HYPERPARAMETER_SPACE)
time_a = time.time()
auto_ml = AutoML(
pipeline,
refit_trial=True,
n_trials=n_iter,
cache_folder_when_no_handle=resumable_pipeline_folder,
validation_splitter=ValidationSplitter(0.2),
hyperparams_optimizer=RandomSearchHyperparameterSelectionStrategy(),
scoring_callback=ScoringCallback(mean_squared_error, higher_score_is_better=False),
callbacks=[
MetricCallback('mse', metric_function=mean_squared_error, higher_score_is_better=False)
]
)
auto_ml = auto_ml.fit(DATA_INPUTS, EXPECTED_OUTPUTS)
outputs = auto_ml.get_best_model().predict(DATA_INPUTS)
time_b = time.time()
pipeline.flush_all_cache()
actual_score = mean_squared_error(EXPECTED_OUTPUTS, outputs)
print('{0} seconds'.format(time_b - time_a))
print('output: {0}'.format(outputs))
print('smallest mse: {0}'.format(actual_score))
print('best hyperparams: {0}'.format(pipeline.get_hyperparams()))
assert isinstance(actual_score, float)
def main(tmpdir, sleep_time: float = 0, n_iter: int = 10):
DATA_INPUTS = np.array(range(100))
EXPECTED_OUTPUTS = np.array(range(100, 200))
HYPERPARAMETER_SPACE = HyperparameterSpace({
'multiplication_1__multiply_by':
RandInt(1, 2),
'multiplication_2__multiply_by':
RandInt(1, 2),
'multiplication_3__multiply_by':
RandInt(1, 2),
})
print('Classic Pipeline:')
pipeline = Pipeline([
('multiplication_1', MultiplyByN()),
('sleep_1', ForEachDataInput(Sleep(sleep_time))),
('multiplication_2', MultiplyByN()),
('sleep_2', ForEachDataInput(Sleep(sleep_time))),
('multiplication_3', MultiplyByN()),
]).set_hyperparams_space(HYPERPARAMETER_SPACE)
time_a = time.time()
best_model = RandomSearch(pipeline,
n_iter=n_iter,
higher_score_is_better=True).fit(
DATA_INPUTS, EXPECTED_OUTPUTS)
outputs = best_model.transform(DATA_INPUTS)
time_b = time.time()
actual_score = mean_squared_error(EXPECTED_OUTPUTS, outputs)
print('{0} seconds'.format(time_b - time_a))
print('output: {0}'.format(outputs))
print('smallest mse: {0}'.format(actual_score))
print('best hyperparams: {0}'.format(pipeline.get_hyperparams()))
assert isinstance(actual_score, float)
print('Resumable Pipeline:')
pipeline = ResumablePipeline(
[('multiplication_1', MultiplyByN()),
('ForEach(sleep_1)', ForEachDataInput(Sleep(sleep_time))),
('checkpoint1', ExpandDim(DefaultCheckpoint())),
('multiplication_2', MultiplyByN()),
('sleep_2', ForEachDataInput(Sleep(sleep_time))),
('checkpoint2', ExpandDim(DefaultCheckpoint())),
('multiplication_3', MultiplyByN())],
cache_folder=tmpdir).set_hyperparams_space(HYPERPARAMETER_SPACE)
time_a = time.time()
best_model = RandomSearch(pipeline,
n_iter=n_iter,
higher_score_is_better=True).fit(
DATA_INPUTS, EXPECTED_OUTPUTS)
outputs = best_model.transform(DATA_INPUTS)
time_b = time.time()
pipeline.flush_all_cache()
actual_score = mean_squared_error(EXPECTED_OUTPUTS, outputs)
print('{0} seconds'.format(time_b - time_a))
print('output: {0}'.format(outputs))
print('smallest mse: {0}'.format(actual_score))
print('best hyperparams: {0}'.format(pipeline.get_hyperparams()))
assert isinstance(actual_score, float)
def main():
boston = load_boston()
X, y = shuffle(boston.data, boston.target, random_state=13)
X = X.astype(np.float32)
X_train, X_test, y_train, y_test = train_test_split(X,
y,
test_size=0.25,
shuffle=False)
# Note that the hyperparameter spaces are defined here during the pipeline definition, but it could be already set
# within the classes ar their definition if using custom classes, or also it could be defined after declaring the
# pipeline using a flat dict or a nested dict.
p = Pipeline([
AddFeatures([
SKLearnWrapper(
PCA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})),
SKLearnWrapper(
FastICA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})),
]),
ModelStacking(
[
SKLearnWrapper(
GradientBoostingRegressor(),
HyperparameterSpace({
"n_estimators": RandInt(50, 600),
"max_depth": RandInt(1, 10),
"learning_rate": LogUniform(0.07, 0.7)
})),
SKLearnWrapper(
KMeans(),
HyperparameterSpace({"n_clusters": RandInt(5, 10)})),
],
joiner=NumpyTranspose(),
judge=SKLearnWrapper(
Ridge(),
HyperparameterSpace({
"alpha": LogUniform(0.7, 1.4),
"fit_intercept": Boolean()
})),
)
])
print("Meta-fitting on train:")
p = p.meta_fit(X_train,
y_train,
metastep=RandomSearch(
n_iter=10,
higher_score_is_better=True,
validation_technique=KFoldCrossValidationWrapper(
scoring_function=r2_score, k_fold=10)))
# Here is an alternative way to do it, more "pipeliney":
# p = RandomSearch(
# p,
# n_iter=15,
# higher_score_is_better=True,
# validation_technique=KFoldCrossValidation(scoring_function=r2_score, k_fold=3)
# ).fit(X_train, y_train)
print("")
print("Transforming train and test:")
y_train_predicted = p.predict(X_train)
y_test_predicted = p.predict(X_test)
print("")
print("Evaluating transformed train:")
score_transform = r2_score(y_train_predicted, y_train)
print('R2 regression score:', score_transform)
print("")
print("Evaluating transformed test:")
score_test = r2_score(y_test_predicted, y_test)
print('R2 regression score:', score_test)
def main(tmpdir):
boston = load_boston()
X, y = shuffle(boston.data, boston.target, random_state=13)
X = X.astype(np.float32)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, shuffle=False)
# Note that the hyperparameter spaces are defined here during the pipeline definition, but it could be already set
# within the classes ar their definition if using custom classes, or also it could be defined after declaring the
# pipeline using a flat dict or a nested dict.
p = Pipeline([
AddFeatures([
SKLearnWrapper(
PCA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})
),
SKLearnWrapper(
FastICA(n_components=2),
HyperparameterSpace({"n_components": RandInt(1, 3)})
),
]),
ModelStacking([
SKLearnWrapper(
GradientBoostingRegressor(),
HyperparameterSpace({
"n_estimators": RandInt(50, 300), "max_depth": RandInt(1, 4),
"learning_rate": LogUniform(0.07, 0.7)
})
),
SKLearnWrapper(
KMeans(),
HyperparameterSpace({"n_clusters": RandInt(5, 10)})
),
],
joiner=NumpyTranspose(),
judge=SKLearnWrapper(
Ridge(),
HyperparameterSpace({"alpha": LogUniform(0.7, 1.4), "fit_intercept": Boolean()})
),
)
])
print("Meta-fitting on train:")
auto_ml = AutoML(
p,
validation_splitter=ValidationSplitter(0.20),
refit_trial=True,
n_trials=10,
epochs=1, # 1 epoc here due to using sklearn models that just fit once.
cache_folder_when_no_handle=str(tmpdir),
scoring_callback=ScoringCallback(mean_squared_error, higher_score_is_better=False),
callbacks=[MetricCallback('mse', metric_function=mean_squared_error, higher_score_is_better=False)],
hyperparams_repository=InMemoryHyperparamsRepository(cache_folder=str(tmpdir))
)
random_search = auto_ml.fit(X_train, y_train)
p = random_search.get_best_model()
print("")
print("Transforming train and test:")
y_train_predicted = p.predict(X_train)
y_test_predicted = p.predict(X_test)
print("")
print("Evaluating transformed train:")
score_transform = r2_score(y_train_predicted, y_train)
print('R2 regression score:', score_transform)
print("")
print("Evaluating transformed test:")
score_test = r2_score(y_test_predicted, y_test)
print('R2 regression score:', score_test)
def main():
def accuracy(data_inputs, expected_outputs):
return np.mean(
np.argmax(np.array(data_inputs), axis=1) == np.argmax(
np.array(expected_outputs), axis=1))
# load the dataset
df = read_csv('data/winequality-white.csv', sep=';')
data_inputs = df.values
data_inputs[:, -1] = data_inputs[:, -1] - 1
n_features = data_inputs.shape[1] - 1
n_classes = 10
p = Pipeline([
TrainOnlyWrapper(DataShuffler()),
ColumnTransformerInputOutput(
input_columns=[(
[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10], ToNumpy(np.float32)
)],
output_columns=[(11, Identity())]
),
OutputTransformerWrapper(PlotDistribution(column=-1)),
MiniBatchSequentialPipeline([
Tensorflow2ModelStep(
create_model=create_model,
create_loss=create_loss,
create_optimizer=create_optimizer
) \
.set_hyperparams(HyperparameterSamples({
'n_dense_layers': 2,
'input_dim': n_features,
'optimizer': 'adam',
'activation': 'relu',
'kernel_initializer': 'he_uniform',
'learning_rate': 0.01,
'hidden_dim': 20,
'n_classes': 3
})).set_hyperparams_space(HyperparameterSpace({
'n_dense_layers': RandInt(2, 4),
'hidden_dim_layer_multiplier': Uniform(0.30, 1),
'input_dim': FixedHyperparameter(n_features),
'optimizer': Choice([
OPTIMIZERS.ADAM.value,
OPTIMIZERS.SGD.value,
OPTIMIZERS.ADAGRAD.value
]),
'activation': Choice([
ACTIVATIONS.RELU.value,
ACTIVATIONS.TANH.value,
ACTIVATIONS.SIGMOID.value,
ACTIVATIONS.ELU.value,
]),
'kernel_initializer': Choice([
KERNEL_INITIALIZERS.GLOROT_NORMAL.value,
KERNEL_INITIALIZERS.GLOROT_UNIFORM.value,
KERNEL_INITIALIZERS.HE_UNIFORM.value
]),
'learning_rate': LogUniform(0.005, 0.01),
'hidden_dim': RandInt(3, 80),
'n_classes': FixedHyperparameter(n_classes)
}))
], batch_size=33),
OutputTransformerWrapper(Pipeline([
ExpandDim(),
OneHotEncoder(nb_columns=n_classes, name='classes')
]))
])
auto_ml = AutoML(
pipeline=p,
hyperparams_repository=InMemoryHyperparamsRepository(
cache_folder='trials'),
hyperparams_optimizer=RandomSearchHyperparameterSelectionStrategy(),
validation_splitter=ValidationSplitter(test_size=0.30),
scoring_callback=ScoringCallback(accuracy,
higher_score_is_better=True),
callbacks=[
MetricCallback(
name='classification_report_imbalanced_metric',
metric_function=classificaiton_report_imbalanced_metric,
higher_score_is_better=True),
MetricCallback(name='f1',
metric_function=f1_score_weighted,
higher_score_is_better=True),
MetricCallback(name='recall',
metric_function=recall_score_weighted,
higher_score_is_better=True),
MetricCallback(name='precision',
metric_function=precision_score_weighted,
higher_score_is_better=True),
EarlyStoppingCallback(max_epochs_without_improvement=3)
],
n_trials=200,
refit_trial=True,
epochs=75)
auto_ml = auto_ml.fit(data_inputs=data_inputs)
def main():
# Define classification models, and hyperparams.
# See also HyperparameterSpace documentation : https://www.neuraxle.org/stable/api/neuraxle.hyperparams.space.html#neuraxle.hyperparams.space.HyperparameterSpace
decision_tree_classifier = SKLearnWrapper(
DecisionTreeClassifier(),
HyperparameterSpace({
'criterion': Choice(['gini', 'entropy']),
'splitter': Choice(['best', 'random']),
'min_samples_leaf': RandInt(2, 5),
'min_samples_split': RandInt(2, 4)
}))
extra_tree_classifier = SKLearnWrapper(
ExtraTreeClassifier(),
HyperparameterSpace({
'criterion': Choice(['gini', 'entropy']),
'splitter': Choice(['best', 'random']),
'min_samples_leaf': RandInt(2, 5),
'min_samples_split': RandInt(2, 4)
}))
ridge_classifier = Pipeline([
OutputTransformerWrapper(NumpyRavel()),
SKLearnWrapper(
RidgeClassifier(),
HyperparameterSpace({
'alpha': Choice([0.0, 1.0, 10.0, 100.0]),
'fit_intercept': Boolean(),
'normalize': Boolean()
}))
]).set_name('RidgeClassifier')
logistic_regression = Pipeline([
OutputTransformerWrapper(NumpyRavel()),
SKLearnWrapper(
LogisticRegression(),
HyperparameterSpace({
'C': LogUniform(0.01, 10.0),
'fit_intercept': Boolean(),
'penalty': Choice(['none', 'l2']),
'max_iter': RandInt(20, 200)
}))
]).set_name('LogisticRegression')
random_forest_classifier = Pipeline([
OutputTransformerWrapper(NumpyRavel()),
SKLearnWrapper(
RandomForestClassifier(),
HyperparameterSpace({
'n_estimators': RandInt(50, 600),
'criterion': Choice(['gini', 'entropy']),
'min_samples_leaf': RandInt(2, 5),
'min_samples_split': RandInt(2, 4),
'bootstrap': Boolean()
}))
]).set_name('RandomForestClassifier')
# Define a classification pipeline that lets the AutoML loop choose one of the classifier.
# See also ChooseOneStepOf documentation : https://www.neuraxle.org/stable/api/neuraxle.steps.flow.html#neuraxle.steps.flow.ChooseOneStepOf
pipeline = Pipeline([
ChooseOneStepOf([
decision_tree_classifier, extra_tree_classifier, ridge_classifier,
logistic_regression, random_forest_classifier
])
])
# Create the AutoML loop object.
# See also AutoML documentation : https://www.neuraxle.org/stable/api/neuraxle.metaopt.auto_ml.html#neuraxle.metaopt.auto_ml.AutoML
auto_ml = AutoML(
pipeline=pipeline,
hyperparams_optimizer=RandomSearchHyperparameterSelectionStrategy(),
validation_splitter=ValidationSplitter(test_size=0.20),
scoring_callback=ScoringCallback(accuracy_score,
higher_score_is_better=True),
n_trials=7,
epochs=1,
hyperparams_repository=HyperparamsJSONRepository(cache_folder='cache'),
refit_trial=True,
continue_loop_on_error=False)
# Load data, and launch AutoML loop !
X_train, y_train, X_test, y_test = generate_classification_data()
auto_ml = auto_ml.fit(X_train, y_train)
# Get the model from the best trial, and make predictions using predict.
# See also predict documentation : https://www.neuraxle.org/stable/api/neuraxle.base.html#neuraxle.base.BaseStep.predict
best_pipeline = auto_ml.get_best_model()
y_pred = best_pipeline.predict(X_test)
accuracy = accuracy_score(y_true=y_test, y_pred=y_pred)
print("Test accuracy score:", accuracy)
shutil.rmtree('cache')
