[liblinear] 01/04: Update man pages

Christian Kastner ckk at moszumanska.debian.org
Tue Sep 15 22:57:12 UTC 2015


This is an automated email from the git hooks/post-receive script.

ckk pushed a commit to branch master
in repository liblinear.

commit cc099663c4953ccb6ad68f2613378f6de96f3644
Author: Christian Kastner <ckk at kvr.at>
Date:   Wed Sep 16 00:39:06 2015 +0200

    Update man pages
---
 debian/liblinear-predict.1 |  3 +++
 debian/liblinear-train.1   | 27 +++++++++++++++++++++++++++
 2 files changed, 30 insertions(+)

diff --git a/debian/liblinear-predict.1 b/debian/liblinear-predict.1
index 7c744ed..2af4ba6 100644
--- a/debian/liblinear-predict.1
+++ b/debian/liblinear-predict.1
@@ -15,6 +15,9 @@ A summary of options is included below.
 .TP
 .B \-b \fI(0|1)\fP
 Whether to output probability estimates or not (default: \fB0\fP)
+.TP
+.B \-q
+quiet mode (no outputs)
 .SH EXAMPLES
 Train a linear SVM using L2-loss function with \fBlinear-train\fP(1):
 .sp
diff --git a/debian/liblinear-train.1 b/debian/liblinear-train.1
index 4beafac..a82b24c 100644
--- a/debian/liblinear-train.1
+++ b/debian/liblinear-train.1
@@ -84,6 +84,9 @@ Weight-adjusts the parameter C of class \fIi\fP by the value \fIweight\fP
 .B \-v \fIn\fP
 \fIn\fP-fold cross validation mode
 .TP
+.B \-C
+Find parameter C (only for \-s 0 and 2)
+.TP
 .B \-q
 Quiet mode (no outputs).
 .SH EXAMPLES
@@ -113,6 +116,30 @@ tolerance 0.001 instead of the default 0.1 for more accurate solutions:
 .fi
 .RE
 .sp
+Conduct cross validation many times by L2-loss SVM and find the parameter
+C which achieves the best cross validation accuracy:
+.sp
+.RS 10
+.nf
+ train \-C datafile
+.fi
+.RE
+.sp
+For parameter selection by \-C, users can specify other
+solvers (currently \-s 0 and \-s 2 are supported) and
+different number of CV folds. Further, users can use
+the \-c option to specify the smallest C value of the
+search range. This setting is useful when users want
+to rerun the parameter selection procedure from a
+specified C under a different setting, such as a stricter
+stopping tolerance \-e 0.0001 in the above example.
+.sp
+.RS 10
+.nf
+ train \-C \-s 0 \-v 3 \-c 0.5 \-e 0.0001 datafile
+.fi
+.RE
+.sp
 Train four classifiers:
 .RS 18
 .sp

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/liblinear.git



More information about the debian-science-commits mailing list