dekker.one
Group work for a Monash Research Methods course
some results
+8
+8
mini_proj/report/waldo.tex
+8
mini_proj/report/waldo.tex
···
297
297
We can see by the results that Deep Neural Networks outperform our benchmark
298
298
classification models, although the time required to train these networks is
299
299
significantly greater.
300
+
301
+
% models that learn relationships between pixels outperform those that don't
302
+
303
+
Of the benchmark classifiers we see the best performance with Random
304
+
Forests and the worst performance with K Nearest Neighbours. As supported
305
+
by the rest of the results, this comes down to a models ability to learn
306
+
the hidden relationships between the pixels. This is made more apparent by
307
+
performance of the Neural Networks.
300
308
301
309
\section{Conclusion} \label{sec:conclusion}
302
310