homework1b solutions

00089 runtime in cpu seconds 060 number of sv 141

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 65 examples read) Setting default regularization parameter C=0.0000 Optimizing...<snip>...done. (163 iterations) Optimization finished (9 misclassified, maxdiff=0.00089). Runtime in cpu-seconds: 0.60 Number of SV: 141 (including 77 at upper bound) L1 loss: loss=35.80801 Norm of weight vector: |w|=0.00365 Norm of longest example vector: |x|=4380.65657 Estimated VCdim of classifier: VCdim<=142.17389 Computing XiAlpha-estimates...done Runtime for XiAlpha-estimates in cpu-seconds: 0.01 XiAlpha-estimate of the error: error<=0.92% (rho=1.00,depth=0) XiAlpha-estimate of the recall: recall=>99.12% (rho=1.00,depth=0) XiAlpha-estimate of the precision: precision=>99.15% (rho=1.00,depth=0) Number of kernel evaluations: 135983 Writing model file...done Cite as: Tommi Jaakkola, course materials for 6.867 Machine Learning, Fall 2006. MIT OpenCourseWare (http://ocw.mit.edu/), Massachusetts Institute of Technology. Downloaded on [DD Month YYYY]. In this case, nine training examples are misclassified. Again, there are many ways of extracting the misclassified images. Here is how it is done with unix power tools: $ svm_classify train-01-images.svm svm_model Reading model...OK. (141 support vectors read) Classifying test examples..<snip>..done Runtime (without IO) in cpu-seconds: 0.00 Accuracy on test set: 99.93% (12656 correct, 9 incorrect, 12665 total) Precision/recall on test set: 99.93%/99.94% $ cat train-01-images.svm | awk ’{ print $1 }’ | paste - svm_predictions \ | nl | awk ’{ if ($2 * $3 < 0) { print $1 } }’ 315 2824 4496 4612 6031 6627 8185 8547 9327 Here are the images: train-0003...
View Full Document

This document was uploaded on 03/20/2014 for the course EECS 6.867 at MIT.

Ask a homework question - tutors are online