def recursive_feature_elimination(input_data,
feature_names,
estimator=SVC(kernel='linear'),
n_features_to_select=None,
step=0.1):
"""
Recursively eliminates features from x_train and x_test using
scikit-learn's RFE, see documentation:
http://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFE.html
If feature_names is given it is also returned with any features from
x_train and x_test also removed from feature_names.
Args:
input_data (tuple): x_train, y_train, x_test, y_test
feature_names (list): The names of all features before
feature selection or None.
estimator (object): Passed to RFE, see documentation
n_features_to_select (int or None): Passed to RFE, see documentation
step (int or float): Passed to RFE, see documentation
Returns:
tuple: (x_train, y_train, x_test, y_test), feature_names, input_Args
"""
x_train = input_data[0]
y_train = input_data[1]
x_test = input_data[2]
y_test = input_data[3]
dims = len(x_train.shape)
if dims == 3:
x_train = flatten(x_train)
x_test = flatten(x_test)
feature_selector = RFE(estimator, n_features_to_select, step)
x_train = feature_selector.fit_transform(x_train, y_train)
x_test = feature_selector.transform(x_test)
if dims == 3:
x_train = make3D(x_train)
x_test = make3D(x_test)
output_data = (x_train, y_train, x_test, y_test)
if feature_names is not None:
mask = feature_selector.get_support()
feature_names = feature_names[mask]
args = {
'estimator': estimator,
'n_features_to_select': n_features_to_select,
'step': step
}
return output_data, feature_names, args
def neural_network(input_data, feature_names=None, validate=True):
"""
Constructs, compiles, trains and tests/makes predictions with a neural
network.
Args:
input_data (tuple): x_train, y_train, x_test, y_test
feature_names (list): Ignored, here for compatability.
validate (bool): If True, an accuracy is returned, if False a
list of predictions for x_test is returned.
Returns:
float: model accuracy
or
list: predicted classifications for x_test.
"""
feature_names = None
x_train = input_data[0]
y_train = input_data[1]
x_test = input_data[2]
y_test = input_data[3]
all_labels = np.concatenate((y_train, y_test))
unique_labels = np.unique(all_labels)
num_classes = unique_labels.shape[0]
y_train = to_categorical(y_train, num_classes=num_classes)
if validate:
y_test = to_categorical(y_test, num_classes=num_classes)
if len(x_train.shape) == 2:
x_train = make3D(x_train)
x_test = make3D(x_test)
model = Sequential()
model.add(
Conv1D(filters=10,
kernel_size=3,
activation='relu',
input_shape=(x_train.shape[1], 1)))
model.add(Flatten())
model.add(Dense(num_classes, activation='softmax'))
model.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
model.fit(x_train, y_train, epochs=50, batch_size=10, verbose=0)
if validate:
evaluation = model.evaluate(x_test, y_test, batch_size=1, verbose=0)
output = evaluation[1]
else:
output = model.predict(x_test)
return (output, feature_names)
def select_percentile(input_data,
feature_names,
score_func=chi2,
percentile=5):
"""
Selects the percentile best features in x_train, removes the rest of the
features from x_train and x_test. Selects the best features by using the
score function score_func and scikit-learn's SelectPercentile. If
feature_names is given it is also returned with any features removed from
x_train and x_test also removed from feature_names.
Args:
input_data (tuple): x_train, y_train, x_test, y_test
feature_names (list): The names of all features before selection or
None.
score_func (function): The score function to be passed to SelectKBest
percentile (int): Percentile of features to keep.
Returns:
tuple: (x_train, y_train, x_test, y_test), feature_names, input_args
"""
if score_func == f_classif:
input_data, feature_names, _ = remove_constant(input_data,
feature_names)
x_train = input_data[0]
y_train = input_data[1]
x_test = input_data[2]
y_test = input_data[3]
dims = len(x_train.shape)
if dims == 3:
x_train = flatten(x_train)
x_test = flatten(x_test)
feature_selector = SelectPercentile(score_func=score_func,
percentile=percentile)
x_train = feature_selector.fit_transform(x_train, y_train)
x_test = feature_selector.transform(x_test)
if dims == 3:
x_train = make3D(x_train)
x_test = make3D(x_test)
output_data = (x_train, y_train, x_test, y_test)
if feature_names is not None:
mask = feature_selector.get_support()
feature_names = feature_names[mask]
return output_data, feature_names, {
'score_func': score_func,
'percentile': percentile
}
def select_fdr(input_data,
feature_names=None,
score_func=f_classif,
alpha=0.05):
if score_func == f_classif:
input_data, feature_names, _ = remove_constant(input_data,
feature_names)
x_train = input_data[0]
y_train = input_data[1]
x_test = input_data[2]
y_test = input_data[3]
dims = len(x_train.shape)
if dims == 3:
x_train = flatten(x_train)
x_test = flatten(x_test)
done = False
increment = alpha
while not done:
feature_selector = SelectFdr(score_func=score_func, alpha=alpha)
temp_x_train = feature_selector.fit_transform(x_train, y_train)
temp_x_test = feature_selector.transform(x_test)
if temp_x_train.shape[1] > 1 and temp_x_test.shape[1] > 1:
done = True
x_train = temp_x_train
x_test = temp_x_test
else:
msg = 'Feature selection was too aggresive, '
msg += 'increasing alpha from {} to {}'.format(
alpha, alpha + increment)
alpha += increment
logging.warning(msg)
if dims == 3:
x_train = make3D(x_train)
x_test = make3D(x_test)
output_data = (x_train, y_train, x_test, y_test)
if feature_names is not None:
mask = feature_selector.get_support()
feature_names = feature_names[mask]
logging.info('Selected {} features'.format(x_train.shape[1]))
final_args = {'score_func': score_func, 'alpha': alpha}
return output_data, feature_names, final_args
def recursive_feature_elimination_cv(input_data,
feature_names,
step=0.1,
cv=3,
estimator=SVC(kernel='linear')):
"""
Recursively elinates features from x_train and x_test with cross
validation, uses scikit-learn's RFECV see documentation:
http://scikit-learn.org/stable/modules/generated/sklearn.feature_selection.RFECV.html
If feature_names is given it is also returned with any features from
x_train and x_test also removed from feature_names.
Args:
input_data (tuple): x_train, y_train, x_test, y_test
feature_names: The names of all features before feature
selection or None.
estimator (object): Passed to RFECV, see documentation
step (int or float): Passed to RFECV, see documentation
cv (int): Passed to RFECV, see documentation
Returns:
tuple: (x_train, y_train, x_test, y_test), feature_names, input_args
"""
x_train = input_data[0]
y_train = input_data[1]
x_test = input_data[2]
y_test = input_data[3]
dims = len(x_train.shape)
if dims == 3:
x_train = flatten(x_train)
x_test = flatten(x_test)
feature_selector = RFECV(estimator, step, cv)
x_train = feature_selector.fit_transform(x_train, y_train)
x_test = feature_selector.transform(x_test)
if dims == 3:
x_train = make3D(x_train)
x_test = make3D(x_test)
output_data = (x_train, y_train, x_test, y_test)
if feature_names is not None:
mask = feature_selector.get_support()
feature_names = feature_names[mask]
args = {'step': step, 'cv': cv, 'estimator': estimator}
return output_data, feature_names, args
def variance_threshold(input_data, feature_names, threshold=0.16):
"""
Removes all features from x_train and x_test whose variances in x_train is
less than threshold. Uses scikit-learn's VarianceThreshold If feature_names
is given it is also returned with any features removed from x_train and
x_test also removed from feature_names.
Args:
input_data (tuple): x_train, y_train, x_test, y_test
feature_names (list): The names of all features before selection or
None.
threshold (float): Lower limit of variance for a feature to be kept
Returns:
tuple: (x_train, y_train, x_test, y_test), feature_names, input_args
"""
x_train = input_data[0]
y_train = input_data[1]
x_test = input_data[2]
y_test = input_data[3]
dims = len(x_train.shape)
if dims == 3:
x_train = flatten(x_train)
x_test = flatten(x_test)
feature_selector = VarianceThreshold(threshold=threshold)
x_train = feature_selector.fit_transform(x_train)
x_test = feature_selector.transform(x_test)
if dims == 3:
x_train = make3D(x_train)
x_test = make3D(x_test)
output_data = (x_train, y_train, x_test, y_test)
if feature_names is not None:
mask = feature_selector.get_support()
feature_names = feature_names[mask]
return output_data, feature_names, {'threshold': threshold}
def setUp(self):
self.input = np.array([[[1], [2], [3]], [[4], [5], [6]], [[7], [8],
[9]]])
self.output = np.array([[1, 2, 3], [4, 5, 6], [7, 8, 9]])
self.flat = flatten(self.input)
self.threeD = make3D(self.output)