def test_svm_classification():
y_signed_train = (y_train * 2) - 1
y_signed_test = (y_test * 2) - 1
for kernel in [RBF(gamma=0.1), Linear()]:
model = SVM(max_iter=250, kernel=kernel)
model.fit(X_train, y_signed_train)
predictions = model.predict(X_test)
assert accuracy(y_signed_test, predictions) >= 0.8
def test_svm_classification():
y_signed_train = (y_train * 2) - 1
y_signed_test = (y_test * 2) - 1
for kernel in [RBF(gamma=0.1), Linear()]:
model = SVM(max_iter=250, kernel=kernel)
model.fit(X_train, y_signed_train)
predictions = model.predict(X_test)
assert accuracy(y_signed_test, predictions) >= 0.8
def addition_nlp(ReccurentLayer):
X_train, X_test, y_train, y_test = addition_dataset(8, 5000)
print(X_train.shape, X_test.shape)
model = NeuralNet(
layers=[
ReccurentLayer,
TimeDistributedDense(1),
Activation('sigmoid'),
],
loss='mse',
optimizer=Adam(),
metric='mse',
batch_size=64,
max_epochs=15,
)
model.fit(X_train, y_train)
predictions = np.round(model.predict(X_test))
predictions = np.packbits(predictions.astype(np.uint8))
y_test = np.packbits(y_test.astype(np.int))
print(accuracy(y_test, predictions))
except ImportError:
from sklearn.cross_validation import train_test_split
from sklearn.datasets import make_classification
from mla.linear_models import LogisticRegression
from mla.metrics import accuracy
from mla.pca import PCA
# logging.basicConfig(level=logging.DEBUG)
# Generate a random binary classification problem.
X, y = make_classification(n_samples=1000, n_features=100, n_informative=75,
random_state=1111, n_classes=2, class_sep=2.5, )
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25,
random_state=1111)
for s in ['svd', 'eigen']:
p = PCA(15, solver=s)
# fit PCA with training data, not entire dataset
p.fit(X_train)
X_train_reduced = p.transform(X_train)
X_test_reduced = p.transform(X_test)
model = LogisticRegression(lr=0.001, max_iters=2500)
model.fit(X_train_reduced, y_train)
predictions = model.predict(X_test_reduced)
print('Classification accuracy for %s PCA: %s'
% (s, accuracy(y_test, predictions)))
def test_knn():
clf = KNN(k=5)
clf.fit(X_train, y_train)
predictions = clf.predict(X_test)
assert accuracy(y_test, predictions) >= 0.95
def test_knn():
clf = KNNClassifier(k=5)
clf.fit(X_train, y_train)
predictions = clf.predict(X_test)
assert accuracy(y_test, predictions) >= 0.95
from sklearn.cross_validation import train_test_split
from sklearn.datasets import make_classification
from mla.linear_models import LogisticRegression
from mla.metrics import accuracy
from mla.pca import PCA
# logging.basicConfig(level=logging.DEBUG)
# Generate a random binary classification problem.
X, y = make_classification(n_samples=1000, n_features=100, n_informative=75,
random_state=1111, n_classes=2, class_sep=2.5, )
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25,
random_state=1111)
for s in ['svd', 'eigen']:
p = PCA(15, solver=s)
# fit PCA with training data, not entire dataset
p.fit(X_train)
X_train_reduced = p.transform(X_train)
X_test_reduced = p.transform(X_test)
model = LogisticRegression(lr=0.001, max_iters=2500)
model.fit(X_train_reduced, y_train)
predictions = model.predict(X_test_reduced)
print('Classification accuracy for %s PCA: %s'
% (s, accuracy(y_test, predictions)))
layers=[
Convolution(n_filters=32,
filter_shape=(3, 3),
padding=(1, 1),
stride=(1, 1)),
Activation('relu'),
Convolution(n_filters=32,
filter_shape=(3, 3),
padding=(1, 1),
stride=(1, 1)),
Activation('relu'),
MaxPooling(pool_shape=(2, 2), stride=(2, 2)),
Dropout(0.5),
Flatten(),
Dense(128),
Activation('relu'),
Dropout(0.5),
Dense(10),
Activation('softmax'),
],
loss='categorical_crossentropy',
optimizer=Adadelta(),
metric='accuracy',
batch_size=128,
max_epochs=3,
)
model.fit(X_train, y_train)
predictions = model.predict(X_test)
print(accuracy(y_test, predictions))
except ImportError:
from sklearn.cross_validation import train_test_split
from sklearn.datasets import make_classification
from mla.linear_models import LogisticRegression
from mla.metrics import accuracy
from mla.pca import PCA
# logging.basicConfig(level=logging.DEBUG)
# Generate a random binary classification problem.
X, y = make_classification(
n_samples=1000, n_features=100, n_informative=75, random_state=1111, n_classes=2, class_sep=2.5
)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=1111)
for s in ["svd", "eigen"]:
p = PCA(15, solver=s)
# fit PCA with training data, not entire dataset
p.fit(X_train)
X_train_reduced = p.transform(X_train)
X_test_reduced = p.transform(X_test)
model = LogisticRegression(lr=0.001, max_iters=2500)
model.fit(X_train_reduced, y_train)
predictions = model.predict(X_test_reduced)
print("Classification accuracy for %s PCA: %s" % (s, accuracy(y_test, predictions)))
y_test = one_hot(y_test.flatten())
print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)
# Approx. 15-20 min. per epoch
model = NeuralNet(
layers=[
Convolution(n_filters=32, filter_shape=(3, 3), padding=(1, 1), stride=(1, 1)),
Activation('relu'),
Convolution(n_filters=32, filter_shape=(3, 3), padding=(1, 1), stride=(1, 1)),
Activation('relu'),
MaxPooling(pool_shape=(2, 2), stride=(2, 2)),
Dropout(0.5),
Flatten(),
Dense(128),
Activation('relu'),
Dropout(0.5),
Dense(10),
Activation('softmax'),
],
loss='categorical_crossentropy',
optimizer=Adadelta(),
metric='accuracy',
batch_size=128,
max_epochs=3,
)
model.fit(X_train, y_train)
predictions = model.predict(X_test)
print(accuracy(y_test, predictions))