Labelled data for P/S separation with CNN

<p>This is the part of training labelled Dataset-B for the publication &quot;P/S separation of multi-component seismic data at land surface based on deep learning&quot;.</p> <p>There are 500-shot data labels:</p> <p>Train/, Val/ &amp; Test/ are the separated file for Training, Validation &amp; Testing</p> <p>The labelled data size are nx*nz=1001*3001 using a IEEE float format</p> <p>One can use the Seisic Unix command to plot for QC: ximage &lt; shotx_mod_1.dat n1=3001 perc=99 &amp;&nbsp;</p> <p>&nbsp;</p> <p>In each directory, the files are named as follows:&nbsp;</p> <p>The horizontal component:</p> <p>&nbsp; &nbsp; shotx_mod_*.dat&nbsp;</p> <p>The vertical component:</p> <p>&nbsp; &nbsp; shotz_mod_*.dat&nbsp;</p> <p>The P-wave label :</p> <p>&nbsp; &nbsp; shotp_*.dat&nbsp;</p> <p>The S-wave label :</p> <p>&nbsp; &nbsp; shots_*.dat&nbsp;</p>