svm-landscape

Langue: en

Version: May 2010 (ubuntu - 24/10/10)

Section: 1 (Commandes utilisateur)

Sommaire

NAME

svm-landscape - Command line interface to svm-landscape in mlpy (version 2.1.0)

SYNOPSIS

svm-landscape [options]

OPTIONS

-h, --help
show this help message and exit
-d FILE, --data=FILE
data - required
-s, --standardize
standardize data
-n, --normalize
normalize data
-k K
k for k-fold cross validation
-c SETS PAIRS
sets and pairs for monte carlo cross validation
-S, --stratified
for stratified cv
-K KERNEL, --kernel=KERNEL
kernel: 'linear', 'gaussian', 'polynomial', 'tr' [default linear]
-P KPARAMETER, --kparameter=KPARAMETER
kernel parameter (two sigma squared) for gaussian and polynomial kernels [default 0.1]
-o COST, --cost=COST
for cost-sensitive classification [-1.0, 1.0] [default 0.0]
-m MIN, --min=MIN
min value for regularization parameter [default -5]
-M MAX, --max=MAX
max value for regularization parameter [default 5]
-p STEPS, --steps=STEPS
steps for regularization parameter [default 11]
-e SCALE, --scale=SCALE
scale for regularization parameter: 'lin' or 'log' [default log]
-l, --lists
Canberra distance indicator