Application of Artificial Neural Networks for the Automatic Spectral Classification

<table class="table-group" id="tab8"><tr><td><table class="table"><tr><td class="thead-hr" colspan="4"><hr/></td></tr><tr class="thead"><td class="align_left">Technique</td><td class="align_center">Mean training RMS</td><td class="align_center">Mean validation RMS</td><td class="align_center">Mean test RMS</td></tr><tr><td class="thead-hr" colspan="4"><hr/></td></tr><tr><td class="align_left">GRNN</td><td class="align_center">74.07</td><td class="align_center">118.49</td><td class="align_center">105.17</td></tr><tr><td class="align_left">LMA</td><td class="align_center">85.84</td><td class="align_center">125.06</td><td class="align_center">121.23</td></tr><tr class="table-tr"><td colspan="4"><hr class="tbody-hr"/></td></tr></table></td></tr></table>

<div>Averaged RMS errors of ANNs using the new LMA proposed classification for LAMOST DR1 spectra.</div>

Mathematical Problems in Engineering

tab8

Table 8

Table 8: Application of Artificial Neural Networks for the Automatic Spectral Classification 