def get_pyrbm(n_features, rng):
pyvotune.dense_input(PyRBMFeatureExtractor)
pyvotune.non_terminal(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='learning_rate',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='momentum',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='l2_weight',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='sparsity',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='scale',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pbool(name='binary', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pbool(name='reconstruction', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(1, 10), name='n_training_epochs',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(1, 20), name='n_gibbs',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(1, 1000), name='batch_size',
rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(5, 1500), name='n_hidden',
rng=rng)(PyRBMFeatureExtractor)
return [PyRBMFeatureExtractor]
def get_theano(n_features, rng):
pyvotune.dense_input(TheanoRBMFeatureExtractor)
pyvotune.non_terminal(TheanoRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='learning_rate', rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(1, 10), name='training_epochs', rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(1, 1000), name='batch_size', rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(1, 1000), name='n_resamples', rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(5, 1500), name='n_hidden', rng=rng)(TheanoRBMFeatureExtractor)
return [TheanoRBMFeatureExtractor]
def get_decomposers(n_features, rng):
pyvotune.dense_input(PCA)
pyvotune.non_terminal(PCA)
pyvotune.pbool(name='whiten', rng=rng)(PCA)
pyvotune.non_terminal(ProjectedGradientNMF)
pyvotune.pint(range=(1, n_features), name='n_components',
rng=rng)(ProjectedGradientNMF)
pyvotune.choice(choices=['nndsvd', 'nndsvda', 'nndsvdar'],
name='init',
rng=rng)(ProjectedGradientNMF)
pyvotune.choice(choices=['data', 'components', None],
name='sparseness',
rng=rng)(ProjectedGradientNMF)
pyvotune.pfloat(range=(1, 10), name='beta', rng=rng)(ProjectedGradientNMF)
pyvotune.pfloat(range=(0.01, 10), name='eta',
rng=rng)(ProjectedGradientNMF)
pyvotune.pint(range=(1, 600), name='max_iter',
rng=rng)(ProjectedGradientNMF)
pyvotune.pint(range=(100, 4000), name='nls_max_iter',
rng=rng)(ProjectedGradientNMF)
# PCA & ProbabilisticPCA are initialized together since ProbabilisticPCA
# wraps PCA
return [PCA, ProbabilisticPCA, ProjectedGradientNMF]
def get_decomposers(n_features, rng):
pyvotune.dense_input(PCA)
pyvotune.non_terminal(PCA)
pyvotune.pbool(name='whiten', rng=rng)(PCA)
pyvotune.non_terminal(ProjectedGradientNMF)
pyvotune.pint(
range=(1, n_features), name='n_components', rng=rng)(ProjectedGradientNMF)
pyvotune.choice(
choices=['nndsvd', 'nndsvda', 'nndsvdar'],
name='init', rng=rng)(ProjectedGradientNMF)
pyvotune.choice(
choices=['data', 'components', None],
name='sparseness', rng=rng)(ProjectedGradientNMF)
pyvotune.pfloat(
range=(1, 10), name='beta', rng=rng)(ProjectedGradientNMF)
pyvotune.pfloat(
range=(0.01, 10), name='eta', rng=rng)(ProjectedGradientNMF)
pyvotune.pint(
range=(1, 600), name='max_iter', rng=rng)(ProjectedGradientNMF)
pyvotune.pint(
range=(100, 4000), name='nls_max_iter', rng=rng)(ProjectedGradientNMF)
# PCA & ProbabilisticPCA are initialized together since ProbabilisticPCA
# wraps PCA
return [PCA, ProbabilisticPCA, ProjectedGradientNMF]
def get_theano(n_features, rng):
pyvotune.dense_input(TheanoRBMFeatureExtractor)
pyvotune.non_terminal(TheanoRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='learning_rate',
rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(1, 10), name='training_epochs',
rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(1, 1000), name='batch_size',
rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(1, 1000), name='n_resamples',
rng=rng)(TheanoRBMFeatureExtractor)
pyvotune.pint(range=(5, 1500), name='n_hidden',
rng=rng)(TheanoRBMFeatureExtractor)
return [TheanoRBMFeatureExtractor]
def get_image_features(n_features, rng):
pyvotune.dense_input(PatchExtractor)
pyvotune.non_terminal(PatchExtractor)
pyvotune.param(
typename="patchsize",
checker_fn=patchsize_checker,
checker_args={
'min_width': 2,
'max_width': 200,
'min_height': 2,
'max_height': 200
},
generator_fn=patchsize_generator,
generator_args={
'min_width': 2,
'max_width': 200,
'min_height': 2,
'max_height': 200
},
name="patch_size")(PatchExtractor)
pyvotune.pint(range=(1, 100), name='max_patches', rng=rng)(PatchExtractor)
return [PatchExtractor]
def get_pyrbm(n_features, rng):
pyvotune.dense_input(PyRBMFeatureExtractor)
pyvotune.non_terminal(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='learning_rate', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='momentum', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='l2_weight', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='sparsity', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pfloat(range=(0., 1.0), name='scale', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pbool(name='binary', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pbool(name='reconstruction', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(1, 10), name='n_training_epochs', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(1, 20), name='n_gibbs', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(1, 1000), name='batch_size', rng=rng)(PyRBMFeatureExtractor)
pyvotune.pint(range=(5, 1500), name='n_hidden', rng=rng)(PyRBMFeatureExtractor)
return [PyRBMFeatureExtractor]
def get_classifiers(n_features, rng):
pyvotune.dense_input(RandomForestClassifier)
pyvotune.excl_terminal(RandomForestClassifier)
pyvotune.choice(
choices=['gini', 'entropy'], name='criterion', rng=rng)(RandomForestClassifier)
pyvotune.pint(range=(1, 1000), name='n_estimators', rng=rng)(RandomForestClassifier)
pyvotune.pfloat(range=(0, 1), name='min_density', rng=rng)(RandomForestClassifier)
pyvotune.pconst(value=1, name='n_jobs', rng=rng)(RandomForestClassifier)
pyvotune.pbool(name='bootstrap', rng=rng)(RandomForestClassifier)
pyvotune.pbool(name='oob_score', rng=rng)(RandomForestClassifier)
pyvotune.dense_input(ExtraTreesClassifier)
pyvotune.excl_terminal(ExtraTreesClassifier)
pyvotune.choice(
choices=['gini', 'entropy'], name='criterion', rng=rng)(ExtraTreesClassifier)
pyvotune.pint(range=(1, 1000), name='n_estimators', rng=rng)(ExtraTreesClassifier)
pyvotune.pfloat(range=(0, 1), name='min_density', rng=rng)(ExtraTreesClassifier)
pyvotune.pconst(value=1, name='n_jobs', rng=rng)(ExtraTreesClassifier)
pyvotune.pbool(name='bootstrap', rng=rng)(ExtraTreesClassifier)
pyvotune.pbool(name='oob_score', rng=rng)(ExtraTreesClassifier)
pyvotune.dense_input(GradientBoostingClassifier)
pyvotune.excl_terminal(GradientBoostingClassifier)
pyvotune.pint(range=(10, 1000), name='n_estimators', rng=rng)(GradientBoostingClassifier)
pyvotune.pint(range=(1, n_features), name='max_features', rng=rng)(GradientBoostingClassifier)
pyvotune.pfloat(range=(0, 1), name='learn_rate', rng=rng)(GradientBoostingClassifier)
pyvotune.pfloat(range=(0, 1), name='subsample', rng=rng)(GradientBoostingClassifier)
pyvotune.dense_input(GaussianNB)
pyvotune.excl_terminal(GaussianNB)
pyvotune.dense_input(MultinomialNB)
pyvotune.pfloat(range=(0, 1), name='alpha', rng=rng)(MultinomialNB)
pyvotune.pbool(name='fit_prior', rng=rng)(MultinomialNB)
pyvotune.excl_terminal(MultinomialNB)
pyvotune.dense_input(KNeighborsClassifier)
pyvotune.excl_terminal(KNeighborsClassifier)
pyvotune.pint(range=(2, 100), name='n_neighbors', rng=rng)(KNeighborsClassifier)
pyvotune.choice(
choices=['uniform', 'distance'], name='weights', rng=rng)(KNeighborsClassifier)
pyvotune.choice(
choices=[
'auto', 'ball_tree', 'kd_tree', 'brute'], name='algorithm', rng=rng)(
KNeighborsClassifier)
pyvotune.pint(range=(10, 100), name='leaf_size', rng=rng)(KNeighborsClassifier)
pyvotune.pint(range=(1, 20), name='p', rng=rng)(KNeighborsClassifier)
pyvotune.excl_terminal(SVC)
pyvotune.pfloat(range=(0, 1000), name='C', rng=rng)(SVC)
pyvotune.choice(choices=['linear', 'poly', 'rbf', 'sigmoid'], name='kernel', rng=rng)(SVC)
pyvotune.pint(range=(0, 10), name='degree', rng=rng)(SVC)
pyvotune.pfloat(range=(0, 100.), name='gamma', rng=rng)(SVC)
pyvotune.pfloat(range=(-10000, 10000.), name='coef0', rng=rng)(SVC)
pyvotune.pbool(name='shrinking', rng=rng)(SVC)
pyvotune.excl_terminal(NuSVC)
pyvotune.pfloat(range=(0, 1.), name='nu', rng=rng)(NuSVC)
pyvotune.choice(choices=['poly', 'rbf', 'sigmoid'], name='kernel', rng=rng)(NuSVC)
pyvotune.pint(range=(0, 10), name='degree', rng=rng)(NuSVC)
pyvotune.pfloat(range=(0, 100.), name='gamma', rng=rng)(NuSVC)
pyvotune.pfloat(range=(-10000, 10000.), name='coef0', rng=rng)(NuSVC)
pyvotune.pbool(name='shrinking', rng=rng)(NuSVC)
return [SVC, NuSVC, GaussianNB, RandomForestClassifier, MultinomialNB,
KNeighborsClassifier, ExtraTreesClassifier,
GradientBoostingClassifier]
def get_regressors(n_features):
pyvotune.dense_input(LinearRegression)
pyvotune.excl_terminal(LinearRegression)
pyvotune.pbool(name='normalize')(LinearRegression)
pyvotune.pbool(name='fit_intercept')(LinearRegression)
pyvotune.dense_input(Ridge)
pyvotune.excl_terminal(Ridge)
pyvotune.pfloat(range=(0, 1), name='alpha')(Ridge)
pyvotune.pbool(name='normalize')(Ridge)
pyvotune.pbool(name='fit_intercept')(Ridge)
pyvotune.dense_input(RandomForestRegressor)
pyvotune.excl_terminal(RandomForestRegressor)
pyvotune.pint(range=(1, 1000), name='n_estimators')(RandomForestRegressor)
pyvotune.pfloat(range=(0, 1), name='min_density')(RandomForestRegressor)
pyvotune.pconst(value=1, name='n_jobs')(RandomForestRegressor)
pyvotune.pbool(name='bootstrap')(RandomForestRegressor)
pyvotune.pbool(name='oob_score')(RandomForestRegressor)
pyvotune.dense_input(ExtraTreesRegressor)
pyvotune.excl_terminal(ExtraTreesRegressor)
pyvotune.pint(range=(1, 1000), name='n_estimators')(ExtraTreesRegressor)
pyvotune.pfloat(range=(0, 1), name='min_density')(ExtraTreesRegressor)
pyvotune.choice(choices=['auto', 'sqrt', 'log2', None], name='max_features')(ExtraTreesRegressor)
pyvotune.pconst(value=1, name='n_jobs')(ExtraTreesRegressor)
pyvotune.pbool(name='bootstrap')(ExtraTreesRegressor)
pyvotune.pbool(name='oob_score')(RandomForestRegressor)
pyvotune.dense_input(GradientBoostingRegressor)
pyvotune.excl_terminal(GradientBoostingRegressor)
pyvotune.pint(range=(10, 1000), name='n_estimators')(GradientBoostingRegressor)
pyvotune.pint(range=(1, n_features), name='max_features')(GradientBoostingRegressor)
pyvotune.pfloat(range=(0, 1), name='learn_rate')(GradientBoostingRegressor)
pyvotune.pfloat(range=(0, 1), name='subsample')(GradientBoostingRegressor)
pyvotune.dense_input(KNeighborsRegressor)
pyvotune.excl_terminal(KNeighborsRegressor)
pyvotune.pint(range=(2, 100), name='n_neighbors')(KNeighborsRegressor)
pyvotune.choice(
choices=['uniform', 'distance'], name='weights')(KNeighborsRegressor)
pyvotune.choice(
choices=[
'auto', 'ball_tree', 'kd_tree', 'brute'], name='algorithm')(
KNeighborsRegressor)
pyvotune.pint(range=(10, 100), name='leaf_size')(KNeighborsRegressor)
pyvotune.pint(range=(1, 20), name='p')(KNeighborsRegressor)
pyvotune.excl_terminal(SVR)
pyvotune.pfloat(range=(0, 1000), name='C')(SVR)
pyvotune.pfloat(range=(0, 2.0), name='epsilon')(SVR)
pyvotune.choice(choices=['linear', 'poly', 'rbf', 'sigmoid'], name='kernel')(SVR)
pyvotune.pint(range=(0, 10), name='degree')(SVR)
pyvotune.pfloat(range=(0, 100.), name='gamma')(SVR)
pyvotune.pfloat(range=(-10000, 10000.), name='coef0')(SVR)
pyvotune.pbool(name='shrinking')(SVR)
pyvotune.excl_terminal(NuSVR)
pyvotune.pfloat(range=(0, 1.), name='nu')(NuSVR)
pyvotune.choice(choices=['poly', 'rbf', 'sigmoid'], name='kernel')(NuSVR)
pyvotune.pint(range=(0, 10), name='degree')(NuSVR)
pyvotune.pfloat(range=(0, 100.), name='gamma')(NuSVR)
pyvotune.pfloat(range=(-10000, 10000.), name='coef0')(NuSVR)
pyvotune.pbool(name='shrinking')(NuSVR)
return [RandomForestRegressor, GradientBoostingRegressor,
KNeighborsRegressor, SVR, NuSVR, ExtraTreesRegressor,
LinearRegression, Ridge]