def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(columns=(7, 8, 11, 12, 13,
14))),
('scale',
preprocessing.StandardScaler(with_mean=True, with_std=True)),
('expand',
preprocessing.PolynomialFeatures(degree=1,
interaction_only=False,
include_bias=False)),
('reduce', decomposition.FastICA(
fun='cube',
random_state=1742,
)),
('select',
feature_selection.SelectKBest(
k=7,
score_func=feature_selection.mutual_info_classif,
)),
('estim', naive_bayes.GaussianNB()),
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
#('kselect', feature_selection.SelectKBest(feature_selection.f_regression, k=115)),
('drop', transformers.ColumnDropper(columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124))),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=True
)),
('select', feature_selection.SelectPercentile(
percentile=85,#59,
score_func=feature_selection.mutual_info_classif
)),
('estim', svm.NuSVC(
nu=0.0525,
kernel='rbf',
gamma='auto',
shrinking=True,
class_weight=None,
random_state=1742
)),
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(
columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124)
)),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=False
)),
('select', feature_selection.SelectPercentile(
percentile=54,
score_func=feature_selection.mutual_info_classif
)),
('estim', semi_supervised.LabelPropagation(
kernel='rbf',
alpha=0.65,
n_neighbors=4,
n_jobs=-1
)),
])
pipe.fit(x, y)
self._transduction = pipe.named_steps['estim'].transduction_
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(
columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124)
)),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=True
)),
('select', feature_selection.SelectPercentile(
percentile=59,#59,
score_func=feature_selection.mutual_info_classif
)),
('select', feature_selection.SelectKBest(
k=101,
score_func=feature_selection.f_classif
)),
('estim', manifold.locally_linear_embedding(
x,
n_neighbors=6,
n_components=101,
eigen_solver='auto',
method='standard'
)),
])
pipe.fit_transform(x)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(
columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124)
)),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=False
)),
('select', feature_selection.SelectPercentile(
percentile=73,
score_func=feature_selection.f_classif
)),
('estim', neighbors.KNeighborsClassifier(
n_neighbors=16,
weights='distance',
metric='euclidean',
n_jobs=-1
))
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop',
transformers.ColumnDropper(columns=(6, 7, 8, 11, 12, 13, 14))),
(
'scale',
preprocessing.StandardScaler(
with_mean=True,
with_std=True # this is not a typo!
)),
('expand',
preprocessing.PolynomialFeatures(degree=2,
interaction_only=False,
include_bias=False)),
('select',
feature_selection.SelectKBest(
k=25, score_func=feature_selection.mutual_info_classif)),
('estim', tree.DecisionTreeClassifier())
#svm.SVC, svm.NuSVC, svm.LinearSVC
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(
columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124)
)),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=False
)),
('reduce', decomposition.FastICA(
n_components=40,
fun='exp',
random_state=1742,
)),
('select', feature_selection.SelectPercentile(
percentile=57,
score_func=feature_selection.mutual_info_classif,
)),
('estim', naive_bayes.GaussianNB()),
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop',
transformers.ColumnDropper(columns=(6, 7, 8, 11, 12, 13, 14))),
(
'scale',
preprocessing.StandardScaler(
with_mean=True,
with_std=False # this is not a typo!
)),
#('scale', preprocessing.RobustScaler(
# with_centering=True, with_scaling=False, quantile_range=(1.0, 99.0)
#)),
('expand',
preprocessing.PolynomialFeatures(degree=2,
interaction_only=False,
include_bias=False)),
('select',
feature_selection.SelectPercentile(
percentile=98, score_func=feature_selection.f_classif)),
('estim',
discriminant_analysis.QuadraticDiscriminantAnalysis(
reg_param=0.0043))
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
# x14 == x10
# x8 == x3
# x9 == x6^2 - C
('drop', transformers.ColumnDropper(columns=(7, 8, 11, 12, 13, 14))
),
('scale',
preprocessing.StandardScaler(with_mean=True, with_std=True)),
('expand',
preprocessing.PolynomialFeatures(degree=2,
interaction_only=True,
include_bias=False)),
('select',
feature_selection.SelectKBest(
k=26, score_func=feature_selection.mutual_info_classif)),
('estim',
neighbors.NearestCentroid(metric='euclidean',
shrink_threshold=None)),
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop',
transformers.ColumnDropper(columns=(6, 7, 8, 11, 12, 13, 14))),
#('select', feature_selection.SelectKBest()),
('scale', preprocessing.StandardScaler()),
('expand', preprocessing.PolynomialFeatures()),
('estim', discriminant_analysis.QuadraticDiscriminantAnalysis()),
])
param_grid = [{
#'select__k': [i for i in range(15, 21)],
#'select__score_func': [feature_selection.f_classif],
'scale__with_mean': [True, False],
'scale__with_std': [True],
'expand__include_bias': [False, True],
'expand__interaction_only': [False, True],
'expand__degree': [1, 2]
#'estim__reg_param': [0.5]
#'estim__alpha': list(0.001 + 1 * i for i in range(0, 5))
}]
grid = model_selection.GridSearchCV(
pipe,
cv=10,
n_jobs=1,
param_grid=param_grid,
verbose=1,
scoring=metrics.make_scorer(metrics.accuracy_score),
)
grid.fit(x, y)
print('Optimal Hyperparametres:')
print('=======================')
for step in grid.best_estimator_.steps:
print(step)
print("CV Score:", grid.best_score_)
self._model = grid.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(
columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124)
)),
('scale', preprocessing.StandardScaler()),
('select', feature_selection.SelectPercentile()),
('estim', discriminant_analysis.QuadraticDiscriminantAnalysis()),
])
param_grid = [{
'scale__with_mean': [True],
'scale__with_std': [True],
#'select__percentile': [i for i in range(40, 81, 3)],
'select__percentile': [i for i in range(40, 51)],
'select__score_func': [
feature_selection.f_classif,
feature_selection.mutual_info_classif
],
'estim__reg_param': [0.1 + 0.025 * i for i in range(-1, 2)]
}]
grid = model_selection.GridSearchCV(
pipe, cv=9, n_jobs=16, param_grid=param_grid, verbose=1,
scoring=metrics.make_scorer(metrics.accuracy_score),
)
grid.fit(x, y)
print('Optimal Hyperparametres:')
print('=======================')
for step in grid.best_estimator_.steps:
print(step)
print("CV Score:", grid.best_score_)
estimator = pipe.named_steps['estim']
if hasattr(estimator, 'transduction_'):
self._transduction = estimator.transduction_
self._model = grid.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(columns=(7, 8, 11, 12, 13,
14))),
('scale',
preprocessing.StandardScaler(with_mean=True, with_std=True)),
('estim',
svm.NuSVC(nu=0.19,
kernel='rbf',
gamma='auto',
shrinking=True,
class_weight='balanced',
random_state=1742)),
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
x, x_val, y, y_val = model_selection.train_test_split(
x, y,
train_size=0.90,
stratify=y,
random_state=2345
)
y = ku.to_categorical(y, num_classes=10)
y_val = ku.to_categorical(y_val, num_classes=10)
model = Sequential()
model.add(Dense(700, input_dim=128, init="he_uniform", activation="relu"))
#model.add(Dropout(0.2))
model.add(Dense(350, init="he_uniform", activation="relu"))
model.add(Dense(10))
model.add(Activation("softmax"))
sgd = SGD(lr=0.1, momentum=0.9, decay=0.0, nesterov=False)
model.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['accuracy'])
history = History()
pipe = pipeline.Pipeline([
#('kselect', feature_selection.SelectKBest(feature_selection.f_regression, k=115)),
('drop', transformers.ColumnDropper(columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124))),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=True
)),
('select', feature_selection.SelectPercentile(
percentile=59,#59,
score_func=feature_selection.mutual_info_classif
)),
('estim', model),
])
pipe.fit(x, y, epochs=50, batch_size=128, callbacks=[history], verbose=2, validation_data=(x_val, y_val), shuffle=2)
self._model = pipe.predict_classes
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(
columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124)
)),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=True
)),
('select', feature_selection.SelectPercentile(
percentile=80,
score_func=feature_selection.f_classif
)),
('estim', discriminant_analysis.LinearDiscriminantAnalysis()),
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(
columns=(0, 3, 5, 14, 26, 35, 40, 65, 72, 95, 99, 104, 124)
)),
('scale', preprocessing.StandardScaler(
with_mean=True,
with_std=True
)),
('select', feature_selection.SelectKBest(
k=101,
score_func=feature_selection.f_classif
)),
('estim', neighbors.NearestCentroid(
metric='euclidean',
shrink_threshold=None
)),
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop', transformers.ColumnDropper(columns=(7, 8, 11, 12, 13,
14))),
('scale',
preprocessing.StandardScaler(with_mean=True, with_std=True)),
('expand',
preprocessing.PolynomialFeatures(degree=1,
interaction_only=False,
include_bias=False)),
('select',
feature_selection.SelectKBest(
k=8, score_func=feature_selection.f_classif)),
('estim',
neighbors.KNeighborsClassifier(n_neighbors=16,
weights='distance',
metric='chebyshev'))
])
pipe.fit(x, y)
self._model = pipe.predict
def _train(self):
x = self._train_features
y = self._train_outputs
pipe = pipeline.Pipeline([
('drop',
transformers.ColumnDropper(columns=(6, 7, 8, 11, 12, 13, 14))),
(
'estim',
ensemble.VotingClassifier(estimators=[
(
'knn',
pipeline.Pipeline([
('scale',
preprocessing.StandardScaler(with_mean=True,
with_std=True)),
('expand',
preprocessing.PolynomialFeatures(
degree=1,
interaction_only=False,
include_bias=False)),
#('select', feature_selection.SelectPercentile(score_func=feature_selection.f_classif)),
('estim', neighbors.KNeighborsClassifier())
])),
('qda',
pipeline.Pipeline(
[('scale',
preprocessing.StandardScaler(
with_mean=True, with_std=False)),
('expand',
preprocessing.PolynomialFeatures(
degree=2,
interaction_only=False,
include_bias=False)),
('select',
feature_selection.SelectPercentile(
score_func=feature_selection.f_classif)),
('estim',
discriminant_analysis.QuadraticDiscriminantAnalysis(
))])),
('dummy',
pipeline.Pipeline([
('estim', dummy.DummyClassifier()),
])),
]))
])
param_grid = [{
#'estim__knn__select__percentile': [i for i in range(5, 8)],
'estim__knn__estim__n_neighbors': [i for i in range(5, 6)],
'estim__knn__estim__weights': ['distance'],
#'estim__knn__estim__metric': ['manhattan', 'euclidean', 'chebyshev'],
'estim__knn__estim__metric': ['euclidean'],
'estim__qda__select__percentile': [i for i in range(94, 95)],
#'estim__qda__estim__reg_param': [0.052 + 0.001 * i for i in range(-5, 6)],
'estim__qda__estim__reg_param': [0.052],
'estim__dummy__estim__strategy': ['most_frequent'],
'estim__dummy__estim__random_state': [1742],
'estim__voting': ['soft'],
'estim__weights': [[8, 8, 5]]
#'estim__weights': list(itertools.product(
# [7.2 + 0.05 * i for i in range(-5, 6)],
# [7.2 + 0.05 * i for i in range(-5, 6)],
# [4.5 + 0.05 * i for i in range(-5, 6)]
#))
}]
grid = model_selection.GridSearchCV(
pipe,
cv=20,
n_jobs=4,
param_grid=param_grid,
verbose=1,
scoring=metrics.make_scorer(metrics.accuracy_score),
)
grid.fit(x, y)
print('Optimal Hyperparametres:')
print('=======================')
for name, step in grid.best_estimator_.steps:
if name == 'estim':
for (name2, _), estim2 in zip(step.estimators,
step.estimators_):
print(' ', name2)
for name3, step3 in estim2.steps:
print(' ', step3)
print('Weights:', step.voting, step.weights)
else:
print(step)
print("CV Score:", grid.best_score_)
self._model = grid.predict
def _train(self):
#Train to distinguish 2 first!
x = self._train_features
y = self._train_outputs
ywith02 = y.copy()
ywith02[ywith02 == 1] = 0
ywith01 = y[y < 2]
xwith01 = x[y < 2]
#First classifier to take out 2
self._classifier1 = pipeline.Pipeline([
('drop',
transformers.ColumnDropper(columns=(6, 7, 8, 11, 12, 13, 14))),
(
'scale',
preprocessing.StandardScaler(
with_mean=True,
with_std=True # this is not a typo!
)),
('expand',
preprocessing.PolynomialFeatures(degree=2,
interaction_only=False,
include_bias=False)),
('select',
feature_selection.SelectKBest(
k=25, score_func=feature_selection.mutual_info_classif)),
('estim',
discriminant_analysis.QuadraticDiscriminantAnalysis(
reg_param=0.0043))
])
#fit all values assuming y=1 is same as y=0
self._classifier1.fit(x, ywith02)
self._classifier2 = pipeline.Pipeline([
('drop', transformers.ColumnDropper(columns=(8, 9, 12, 13, 14,
15))),
(
'scale',
preprocessing.StandardScaler(
with_mean=True,
with_std=True # this is not a typo!
)),
('expand',
preprocessing.PolynomialFeatures(degree=2,
interaction_only=False,
include_bias=False)),
('select',
feature_selection.SelectKBest(
k=25, score_func=feature_selection.mutual_info_classif)),
('estim',
discriminant_analysis.QuadraticDiscriminantAnalysis(
reg_param=0.0043))
])
self._classifier2.fit(xwith01, ywith01)
self._model = self._hierarchical_model