this is for holding javascript data
Roland Szabo edited experiments.tex
almost 10 years ago
Commit id: 4d372eb2cc05ca538e81161e987fca0f888a9cd3
deletions | additions
diff --git a/experiments.tex b/experiments.tex
index 269e879..2a50c20 100644
--- a/experiments.tex
+++ b/experiments.tex
...
All experiments were run multiple types, with the dataset being shuffled each time. In the case of the Random Forests, the multiple runs of the experiments are necessary because the splitting points for the trees and the dataset splits are chosen randomly across runs.
\subsection{Experiments}
For both tasks, the parameters for the algorithms were selected using cross-validation. In the case of the SVM, the search space was on logarithmic scale from $10^{-2}$ to $10^4$ for the regularization parameter. In the case of the random forest, the number of trees used ranged from
50 150 to
300 250, in steps of 50,
and the
maximum depth number of
features to be sampled at each
tree point varied from
10 to 100, in steps of 50, and using the
measure of square root, the
quality of the split was either the Gini impurity measure base 2 logarithm, 10\% or
30\% of the
information gain. total number of features.
Table \ref{table:recog_values} contains the average, maximum and minimum values obtained for the accuracy of the character recognition problem.