The following node is available in the Open Source KNIME predictive analytics and data mining platform version 2.7.1. Discover over 1000 other nodes, as well as enterprise functionality at
http://knime.com.
LIBSVMLearner
LIBSVM v2.89 is an integrated software for support vector classification.
For a more detailed description - especially of the parameters - see
the webpage.
Dialog Options
- Type of SVM
- Set type of SVM (default is C-SVM)
C-SVC, nu-SVC, one-class SVM, epsilon-SVR, nu-SVR.
- Kernel
- Set type of kernel function (default is polynomial).
- linear: u'*v
- polynomial: (gamma*u'*v + coef0)^degree
- radial basis function: exp(-gamma*|u-v|^2)
- sigmoid: tanh(gamma*u'*v + coef0)
- Degree
- Set degree in kernel function.
- Gamma
- Set gamma in kernel function (a good value is 1/nrAttributes).
- Coef0
- Set coef0 in kernel function.
- Cost
- Set the parameter C of C-SVC, epsilon-SVR, and nu-SVR.
- Nu
- Set the parameter nu of nu-SVC, one-class SVM, and nu-SVR.
- Loss-Epsilon
- Set the epsilon in loss function of epsilon-SVR.
- Cachesize
- Set cache memory size in MB.
- Epsilon
- Set tolerance of termination criterion.
- Shrinking
- Whether to use the shrinking heuristics.
- Probability estimates
- Whether to train a SVC or SVR model for probability estimates.
- Target column
- The target column, can be nominal for classification or numerical for regression.
Ports
Input Ports
0 |
DataTable containing the training data. Keep in mind
that the computational complexitiy is in O(n^2), therefore do not use
more than 15.000 rows. |
Output Ports
0 |
The trained SVM model in LIBSVM format. |
This node is contained in KNIME LIBSVM Nodes
provided by KNIME GmbH, Konstanz, Germany.