<div dir="ltr">Primarily bug fixes, especially that the BST was not showing hits for non-GEMC files. Available as show at the end of this missive.<div><div><br></div><div>There is a new menu "FastMC" that doesn't do much yet. Under the "Events" menu you can change the source of events from EVIO to Lund files. Then under FastMC you can select a Lund file and scroll through the events. This will swim the tracks and, using the common FastMC/Geometry methods, will determine the "perfect" hits in DC, and eventually in all detectors.</div><div><br></div><div>We will be using this capability soon to try some proof of concept machine learning. We will train a neural net to give us a guess at p,theta,phi given the DC hits. We will also build a dictionary for quick lookup. </div><div><br></div><div>If it works, there will be celebration. If the neural network can't even work on perfect data, there will be weeping and gnashing of teeth.</div><div><br></div><div>cheers,</div><div>dph</div><div><br></div><div><br></div><div><img src="cid:ii_1553599537449223" alt="Inline image 2" width="462" height="255"><br></div><div><br></div><div><br></div><div><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr">David P. Heddle, Ph.D.<br>Professor of Physics<br>
Christopher Newport University<br>
Newport News, VA 23606<div><br></div><div>757.594.8434 (CNU)</div></div></div></div></div>
</div></div></div>