ksvmlib 0.0.3

Kernel SVM library based on sklearn and GPlib.

Kernel SVM library based on sklearn and GPlib. Provides similar functionality to GPlib for SVMs.

Setup kSVMlib

• Create and activate virtualenv (for python2) or venv (for python3)

# for python3
python3 -m venv .env
# or for python2
python2 -m virtualenv .env

source .env/bin/activate

• Upgrade pip

python -m pip install --upgrade pip

• Install kSVMlib package

python -m pip install ksvmlib

Use kSVMlib

• Import kSVMlib to use it in your python script.

import ksvmlib

• Generate some random data.

import numpy as np
data = {}
data['X'] = np.vstack((
    np.random.multivariate_normal([1, 1], [[1, 0], [0, 1]], 100),
    np.random.multivariate_normal([3, 3], [[1, 0], [0, 1]], 100)
))
data['Y'] = np.vstack((
    np.ones((100, 1)),
    np.zeros((100, 1)),
))

validation = ksvmlib.dm.RandFold(fold_len=0.2, n_folds=1)
train_set, test_set = validation.get_folds(data)[0]

• Initialize the KSVM model and a metric to measure the results.

model = ksvmlib.KSVM(ksvmlib.ker.SquaredExponential())
accuracy = ksvmlib.me.Accuracy()

• Fit the model to the data.

fitting_method = ksvmlib.fit.GridSearch(
    obj_fun=accuracy.fold_measure,
    max_fun_call=300
)
train_validation = ksvmlib.dm.RandFold(fold_len=0.2, n_folds=3)

log = fitting_method.fit(model, train_validation.get_folds(
    train_set
))
print("Fitting log: {}".format(log))

• Finally plot the results.

print("Accuracy: {}".format(accuracy.measure(model, train_set, test_set)))
ksvmlib.plot.kernel_sort_data(model, test_set)

• There are more examples in examples/ directory. Check them out!

Develop kSVMlib

• Download the repository using git

git clone https://gitlab.com/ibaidev/ksvmlib.git
cd ksvmlib
git config user.email 'MAIL'
git config user.name 'NAME'
git config credential.helper 'cache --timeout=300'
git config push.default simple

• Update API documentation

source ./.env/bin/activate
pip install Sphinx
cd docs/
sphinx-apidoc -f -o ./ ../ksvmlib