summaryrefslogtreecommitdiff
path: root/sampling_alg_lac2020/LAC-20.tex
diff options
context:
space:
mode:
authorAndré Nusser <andre.nusser@googlemail.com>2020-02-10 00:33:00 +0100
committerAndré Nusser <andre.nusser@googlemail.com>2020-02-10 00:33:00 +0100
commit10ed729f56a42e3701bc4c33fe6e9cc1bdcdbe54 (patch)
treec8c56009d88958418723030debcdde214c660996 /sampling_alg_lac2020/LAC-20.tex
parent7f8a99795f81eb8659e347764295887232d0dc4b (diff)
Add experiment for number of evaluations.
Diffstat (limited to 'sampling_alg_lac2020/LAC-20.tex')
-rw-r--r--sampling_alg_lac2020/LAC-20.tex28
1 files changed, 21 insertions, 7 deletions
diff --git a/sampling_alg_lac2020/LAC-20.tex b/sampling_alg_lac2020/LAC-20.tex
index 3b55bc8..cbbcb17 100644
--- a/sampling_alg_lac2020/LAC-20.tex
+++ b/sampling_alg_lac2020/LAC-20.tex
@@ -617,13 +617,27 @@ The second experiment we conducted for two different MIDI velocities: MIDI veloc
% \todoandre{Also do an experiment regarding the adaptive search}
To also get an idea of the performance of the new sampling algorithm, we want to see how many power values of samples are evaluated per query.
-Without the smart search optimization described at the end of Section \ref{sec:implementation}, this number would always be the number of samples. However, we expect that the smart search optimization significantly reduces the number of evaluations. To test this hypothesis, we take the above experiment and look at the number of evaluations. You can see the histogram in Figure \ref{fig:evaluations_histogram}. \todo{fix}
-
-\begin{figure}
- % \includegraphics[width=.8\linewidth]{figures/evaluations_histogram.pdf}
- \caption{This plot shows the histogram of the number of evaluations of power values for the queries of experiment bla.\todo{insert correct information}}
- \label{fig:evaluations_histogram}
-\end{figure}
+Without the smart search optimization described at the end of Section \ref{sec:implementation}, this number would always be the number of samples. However, we expect that the smart search optimization significantly reduces the number of evaluations. To test this hypothesis, we take the above experiment and look at the number of evaluations. You can see some data regarding the number of evaluations in Table \ref{tab:evaluation_count}.
+
+% \begin{figure}
+% % \includegraphics[width=.8\linewidth]{figures/evaluations_histogram.pdf}
+% \caption{This plot shows the histogram of the number of evaluations of power values for the queries of experiment bla.\todo{insert correct information}}
+% \label{fig:evaluations_histogram}
+% \end{figure}
+
+\begin{table}
+\caption{Number of evaluations per query.}
+\label{tab:evaluation_count}
+\centering
+\begin{tabular}{|l||rrrrr|}
+\hline
+experiment & sweep & 16 & 48 & 80 & 112 \\
+\hline
+mean evaluations & 6.81 & 13.99 & 12.93 & 10.88 & 4.00 \\
+variance evaluations & 6.09 & 0.04 & 2.34 & 0.25 & 0.00 \\
+\hline
+\end{tabular}
+\end{table}
\todoandre{Summarize experiments}