Questions tagged [svmlight]

SVMLight is an implementation of a Support Vector Machine - supervised learning models with associated learning algorithms that analyze data and recognize patterns, used for classification and regression analysis.

The main features of the program are the the following:

  • fast optimization algorithm
    • working set selection based on steepest feasible descent
    • "shrinking" heuristic
    • caching of kernel evaluations
    • use of folding in the linear case
  • solves classification and regression problems. For multivariate and structured outputs use SVMstruct.
  • solves ranking problems (e. g. learning retrieval functions in STRIVER search engine).
  • computes XiAlpha-estimates of the error rate, the precision, and the recall
  • efficiently computes Leave-One-Out estimates of the error rate, the precision, and the recall
  • includes algorithm for approximately training large transductive SVMs (TSVMs) (see also Spectral Graph Transducer)
  • can train SVMs with cost models and example dependent costs
  • allows restarts from specified vector of dual variables
  • handles many thousands of support vectors
  • handles several hundred-thousands of training examples
  • supports standard kernel functions and lets you define your own
  • uses sparse vector representation
98 questions
0
votes
0 answers

Error in loading SVM Learning Configuration File in Gate Tool

I am new to machine learning. When I tried learning through gate, it is showing some error. The learning configuration file is given below.
user3251664
  • 441
  • 2
  • 7
  • 11
0
votes
1 answer

error loading svm learning configuration file in gate tool

I am new to machine learning. When I tried learning through gate, it is showing some error. The learning configuration file is given below.
user3243366
  • 75
  • 1
  • 1
  • 4
0
votes
0 answers

OpenCV Bad trained Hog SVM

I found this repository for training a SVM with HOG features. I changed the code so the hog window size is not 64x128 but 64x64 instead as I don't want to detect people. My goal is to detect a car steering wheel. After training I have: True…
dephinera
  • 3,703
  • 11
  • 41
  • 75
0
votes
1 answer

SVM Rank regularization parameter

When learning a ranking function with SVM Rank, should the values of regularization parameter C be the same as the ones used in classification? The reason why I am asking is that SVM Rank website suggests that the parameter should be C = C_light *…
zpavlinovic
  • 1,507
  • 1
  • 17
  • 36
0
votes
2 answers

SVM rank works only on tiny datasets

I am using svm-rank. When running svm_rank_learn on a tiny dataset: Training set properties: 3 features, 12 rankings, 596 examples The run finishes in a few seconds and I get a valid model. But when I use a bit larger dataset: Training set…
zvisofer
  • 1,346
  • 18
  • 41
0
votes
1 answer

Load svmlight format error

When I try to use the svmlight python package with data I already converted to svmlight format I get an error. It should be pretty basic, I don't understand what's happening. Here's the code: import svmlight training_data = open('thedata',…
PF_learning
  • 25
  • 13
0
votes
1 answer

using svmlight model file in opencv

I have been working on training pedestrian detection classifier based on HOG features. Presently I have done the followings: a) Extracted HOG features of all files i.e. Positive and Negative and saved those features with label i.e. +1 for positive…
Lahori
  • 1,071
  • 11
  • 20
0
votes
1 answer

Arff to SVM Light

I have created a weka arff, which I'd like to convert to SVMLight format. This is so I can load the arff in MLComp. How do I go about converting an arff?
Adam_G
  • 7,337
  • 20
  • 86
  • 148
0
votes
1 answer

Why can't my Java program find svm_learn or svm_classify?

I am using ClearTK along with SVM-Light programmatically within Java. I have downloaded the package both for ClearTK integration as well as for SVM-Light itself. Whether I run via Eclipse or the command line, I keep getting Cannot find file…
demongolem
  • 9,474
  • 36
  • 90
  • 105
0
votes
1 answer

Is there a way to convert discriminant values in SVMLight multi-class classification into probability scores

I am using SVM Light multi-class classifier for training a classifier with four classes. In the classification stage the classifier outputs the predicted label and the scores for the 4 classes. As the SVM Light website says, these scores are "the…
ritesh
  • 229
  • 1
  • 5
  • 12
0
votes
1 answer

How to get training accuracy in svmlight with cross validation

I want to run cross validation on my training set using SVMlight. It seems that the option for this is -x 1 (although I'm not sure how many folds it implements...). The output is: XiAlpha-estimate of the error: error<=31.76%…
Cheshie
  • 2,777
  • 6
  • 32
  • 51
0
votes
1 answer

SVM vector of weights

I have a classification task, and I use svm_perf application. The question is having trained the model I wonder whether it's possible to get the weight of the features. There is an -a parametes which outputs the alphas, honestly I don't recall…
user16168
  • 625
  • 2
  • 8
  • 16
0
votes
1 answer

Efficient multiclass SVM package

Do you guys have experience with an efficient multiclass SVM package? I am currently using libsvm, but on 300k instances and 200 features it takes a very long time (24+ hours) to train. Thanks!
langLord
  • 21
  • 4
0
votes
0 answers

Input file format for email classification with svm-light

I am working with email subject, so I have 20 emails i want to classify, and a file with 20 lines - one line has one email subject.I have been working on it, but I am unable to figure out what the features refer to and the format of the input file…
student001
  • 533
  • 1
  • 7
  • 20
0
votes
1 answer

Is there a scaling function in SVMlight?

The LIBSVM guide states that data sets should be scaled to a certain range before running SVM on them. In Appendix B of the guide, they show a LIBSVM function svm-scale that does the scaling. Is there an equivalent function for SVMlight?
Cheshie
  • 2,777
  • 6
  • 32
  • 51