Got it. Thank you for the explanation. Kind regards, Akash
On Thu, Apr 18, 2019, 11:39 AM Christoph Junghans <[email protected]> wrote: > On Thu, Apr 18, 2019 at 8:01 AM Akash Banerjee <[email protected]> > wrote: > > > > Thanks for the response. Sometimes its in the initial step, and > sometimes it occurs after a couple of iterations. I will try to use the > option as instructed. > > Can you please explain how do I get an idea of the number of points to > be used with the option -inverse.post_update_options.extrapolate.points. > > As of now , what I understand when you say points is the <step> option. > I am guessing lower values of steps would generate a higher number of > points between > > <max> and <min>. Is this correct? > <step> is the spacing between <min> and <max>. <extrapolate.points> is > the number of point, starting from <min> and <max> respectively used > to calculate the average slot used for the extrapolation. Something > like <min>+<points>*<step>. > > You can find the details here: > > https://github.com/votca/csg/blob/master/share/scripts/inverse/table_extrapolate.pl#L129 > > > > > Specifically for extrapolation, I have an additional question. I use the > target RDF to set a <max> and <min>. These max and min values correspond to > the final non-zero value on each side > > of the RDF. Now, to extrapolate correctly, do I increase the max value > and decrease the min value (this would broaden the sampling space for the > RDF) ? Or is extrapolation not at all connected to the <max> and <min> > terms? Please help me understand the above idea. > There are connected as the extrapolation starts at <min> (and <max>). > <min> and <max> basically defines the range, where IBI or some other > method is performed. If you increase the range too much, you risk to > reach areas with poor sampling, i.e. some iteration there is a hit, > some iteration there isn't and that will screw up the update as for > e.g. IBI. > > Christoph > > > > > Thanks for your help! > > Kind regards > > Akash > > > > On Mon, Apr 15, 2019 at 3:33 PM Christoph Junghans <[email protected]> > wrote: > >> > >> On Mon, Apr 15, 2019 at 10:10 AM Akash Banerjee <[email protected]> > wrote: > >> > > >> > To add, I was earlier advised to check the extrapolation for the > input file. If I understand properly, the extrapolation is the max_limit > value. I had extended the min and max limits from 0.2 - 0.4 to 0 - 1.0. > >> > This didint really help. Is there any other way to control > extrapolation? > >> Is it in the initial step or during the iteration? > >> > >> In the iteration you can tweak the number of points to average over: > >> > https://github.com/votca/csg/blob/master/share/scripts/inverse/postupd_extrapolate.sh#L33 > >> with the inverse.post_update_options.extrapolate.points option. > >> > >> We could expose more options to the user if needed: > >> > https://github.com/votca/csg/blob/master/share/scripts/inverse/potential_extrapolate.sh#L25 > >> > >> Christoph > >> > >> > > >> > On Monday, April 15, 2019 at 12:05:09 PM UTC-4, Akash Banerjee wrote: > >> >> > >> >> Dear Votca Developers , > >> >> > >> >> I just want some help to understand the following issue. > >> >> > >> >> No valid value found in bond_2.dpot.pure_ibi at > /usr/share/votca/scripts/inverse/potential_shift.pl line 87. > >> >> > >> >> Please advice me on what I can do to resolve the above. > >> >> > >> >> Thank you for your help! > >> >> > >> >> Kind Regards, > >> >> Akash > >> > > >> > -- > >> > You received this message because you are subscribed to the Google > Groups "votca" group. > >> > To unsubscribe from this group and stop receiving emails from it, > send an email to [email protected]. > >> > To post to this group, send email to [email protected]. > >> > Visit this group at https://groups.google.com/group/votca. > >> > For more options, visit https://groups.google.com/d/optout. > >> > >> > >> > >> -- > >> Christoph Junghans > >> Web: http://www.compphys.de > >> > >> -- > >> You received this message because you are subscribed to the Google > Groups "votca" group. > >> To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected]. > >> To post to this group, send email to [email protected]. > >> Visit this group at https://groups.google.com/group/votca. > >> For more options, visit https://groups.google.com/d/optout. > > > > > > > > -- > > akash > > > > -- > > You received this message because you are subscribed to the Google > Groups "votca" group. > > To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected]. > > To post to this group, send email to [email protected]. > > Visit this group at https://groups.google.com/group/votca. > > For more options, visit https://groups.google.com/d/optout. > > > > -- > Christoph Junghans > Web: http://www.compphys.de > > -- > You received this message because you are subscribed to the Google Groups > "votca" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > Visit this group at https://groups.google.com/group/votca. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "votca" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/votca. For more options, visit https://groups.google.com/d/optout.
