How to reproduce the CRS of test 2?

[xiaowu759]

I used the default setting in inp.dat to test the test 2 in the tests directory. I found that the DRS could be reproduced successfully. But the CRS (say H.dat) exhibits only bimodal not four modes as expected. How to reproduce the CRS in Figure 2b in the paper [Macromol. Theory Simul. 2019, 1900005]? Could you give me some hints, please?

H.dat:
2.078796e-04 -5.785635e+00
2.410452e-04 -5.642137e+00
2.795021e-04 -5.498639e+00
3.240945e-04 -5.355141e+00
3.758014e-04 -5.211644e+00
4.357576e-04 -5.068146e+00
5.052794e-04 -4.924649e+00
5.858929e-04 -4.781154e+00
6.793676e-04 -4.637662e+00
7.877556e-04 -4.494175e+00
9.134360e-04 -4.350699e+00
1.059168e-03 -4.207240e+00
1.228150e-03 -4.063808e+00
1.424092e-03 -3.920420e+00
1.651295e-03 -3.777099e+00
1.914746e-03 -3.633878e+00
2.220230e-03 -3.490805e+00
2.574450e-03 -3.347942e+00
2.985184e-03 -3.205373e+00
3.461448e-03 -3.063208e+00
4.013695e-03 -2.921591e+00
4.654050e-03 -2.780698e+00
5.396568e-03 -2.640755e+00
6.257550e-03 -2.502032e+00
7.255894e-03 -2.364857e+00
8.413517e-03 -2.229617e+00
9.755829e-03 -2.096761e+00
1.131230e-02 -1.966798e+00
1.311709e-02 -1.840299e+00
1.520982e-02 -1.717885e+00
1.763643e-02 -1.600220e+00
2.045019e-02 -1.487988e+00
2.371286e-02 -1.381879e+00
2.749607e-02 -1.282554e+00
3.188286e-02 -1.190621e+00
3.696952e-02 -1.106600e+00
4.286773e-02 -1.030895e+00
4.970695e-02 -9.637634e-01
5.763732e-02 -9.052949e-01
6.683291e-02 -8.554025e-01
7.749559e-02 -8.138203e-01
8.985943e-02 -7.801143e-01
1.041958e-01 -7.537056e-01
1.208195e-01 -7.339024e-01
1.400953e-01 -7.199385e-01
1.624464e-01 -7.110155e-01
1.883635e-01 -7.063446e-01
2.184155e-01 -7.051847e-01
2.532620e-01 -7.068740e-01
2.936680e-01 -7.108541e-01
3.405205e-01 -7.166850e-01
3.948480e-01 -7.240518e-01
4.578430e-01 -7.327624e-01
5.308884e-01 -7.427391e-01
6.155876e-01 -7.540035e-01
7.137999e-01 -7.666547e-01
8.276812e-01 -7.808445e-01
9.597314e-01 -7.967449e-01
1.112849e+00 -8.145124e-01
1.290396e+00 -8.342457e-01
1.496269e+00 -8.559398e-01
1.734987e+00 -8.794369e-01
2.011791e+00 -9.043774e-01
2.332757e+00 -9.301558e-01
2.704931e+00 -9.558872e-01
3.136482e+00 -9.803907e-01
3.636884e+00 -1.002197e+00
4.217121e+00 -1.019589e+00
4.889930e+00 -1.030674e+00
5.670081e+00 -1.033499e+00
6.574700e+00 -1.026205e+00
7.623643e+00 -1.007219e+00
8.839937e+00 -9.754756e-01
1.025028e+01 -9.306622e-01
1.188564e+01 -8.734646e-01
1.378190e+01 -8.057913e-01
1.598070e+01 -7.309370e-01
1.853029e+01 -6.536395e-01
2.148666e+01 -5.799788e-01
2.491470e+01 -5.170739e-01
2.888965e+01 -4.725616e-01
3.349877e+01 -4.538915e-01
3.884324e+01 -4.675359e-01
4.504039e+01 -5.182642e-01
5.222624e+01 -6.086473e-01
6.055853e+01 -7.389039e-01
7.022019e+01 -9.071082e-01
8.142328e+01 -1.109675e+00
9.441375e+01 -1.341978e+00
1.094767e+02 -1.598965e+00
1.269429e+02 -1.875654e+00
1.471957e+02 -2.167486e+00
1.706797e+02 -2.470512e+00
1.979103e+02 -2.781477e+00
2.294854e+02 -3.097803e+00
2.660980e+02 -3.417531e+00
3.085519e+02 -3.739242e+00
3.577791e+02 -4.061963e+00
4.148600e+02 -4.385089e+00
4.810477e+02 -4.708303e+00

[shane5ul]

The easiest way here is to pick a lower lambda. You can do this, for example, by setting: lamC : 0 SmFacLam : -0.8 I modified (improved, I hope) the CRS algorithm after the paper was published to better approximate experimental spectra. I now use a Bayesian formulation to determine the critical lambda since the old method (L-curve) used in the paper often resulted in too small lambda. While that was a good choice for artificial tests such as test2, it created squiggles for more realistic data (see test6, and discussion in the paper). With the new method for critical lambda, the program produces a smooth CRS for test6. I hope this helps.

…

On Mon, Sep 16, 2019 at 2:26 AM xiaowu759 ***@***.***> wrote: I used the default setting in inp.dat to test the test 2 in the tests directory. I found that the DRS could be reproduced successfully. But the CRS (say H.dat) exhibits only bimodal not four modes as expected. How to reproduce the CRS in Figure 2b in the paper [Macromol. Theory Simul. 2019, 1900005]? Could you give me some hints, please? H.dat: 2.078796e-04 -5.785635e+00 2.410452e-04 -5.642137e+00 2.795021e-04 -5.498639e+00 3.240945e-04 -5.355141e+00 3.758014e-04 -5.211644e+00 4.357576e-04 -5.068146e+00 5.052794e-04 -4.924649e+00 5.858929e-04 -4.781154e+00 6.793676e-04 -4.637662e+00 7.877556e-04 -4.494175e+00 9.134360e-04 -4.350699e+00 1.059168e-03 -4.207240e+00 1.228150e-03 -4.063808e+00 1.424092e-03 -3.920420e+00 1.651295e-03 -3.777099e+00 1.914746e-03 -3.633878e+00 2.220230e-03 -3.490805e+00 2.574450e-03 -3.347942e+00 2.985184e-03 -3.205373e+00 3.461448e-03 -3.063208e+00 4.013695e-03 -2.921591e+00 4.654050e-03 -2.780698e+00 5.396568e-03 -2.640755e+00 6.257550e-03 -2.502032e+00 7.255894e-03 -2.364857e+00 8.413517e-03 -2.229617e+00 9.755829e-03 -2.096761e+00 1.131230e-02 -1.966798e+00 1.311709e-02 -1.840299e+00 1.520982e-02 -1.717885e+00 1.763643e-02 -1.600220e+00 2.045019e-02 -1.487988e+00 2.371286e-02 -1.381879e+00 2.749607e-02 -1.282554e+00 3.188286e-02 -1.190621e+00 3.696952e-02 -1.106600e+00 4.286773e-02 -1.030895e+00 4.970695e-02 -9.637634e-01 5.763732e-02 -9.052949e-01 6.683291e-02 -8.554025e-01 7.749559e-02 -8.138203e-01 8.985943e-02 -7.801143e-01 1.041958e-01 -7.537056e-01 1.208195e-01 -7.339024e-01 1.400953e-01 -7.199385e-01 1.624464e-01 -7.110155e-01 1.883635e-01 -7.063446e-01 2.184155e-01 -7.051847e-01 2.532620e-01 -7.068740e-01 2.936680e-01 -7.108541e-01 3.405205e-01 -7.166850e-01 3.948480e-01 -7.240518e-01 4.578430e-01 -7.327624e-01 5.308884e-01 -7.427391e-01 6.155876e-01 -7.540035e-01 7.137999e-01 -7.666547e-01 8.276812e-01 -7.808445e-01 9.597314e-01 -7.967449e-01 1.112849e+00 -8.145124e-01 1.290396e+00 -8.342457e-01 1.496269e+00 -8.559398e-01 1.734987e+00 -8.794369e-01 2.011791e+00 -9.043774e-01 2.332757e+00 -9.301558e-01 2.704931e+00 -9.558872e-01 3.136482e+00 -9.803907e-01 3.636884e+00 -1.002197e+00 4.217121e+00 -1.019589e+00 4.889930e+00 -1.030674e+00 5.670081e+00 -1.033499e+00 6.574700e+00 -1.026205e+00 7.623643e+00 -1.007219e+00 8.839937e+00 -9.754756e-01 1.025028e+01 -9.306622e-01 1.188564e+01 -8.734646e-01 1.378190e+01 -8.057913e-01 1.598070e+01 -7.309370e-01 1.853029e+01 -6.536395e-01 2.148666e+01 -5.799788e-01 2.491470e+01 -5.170739e-01 2.888965e+01 -4.725616e-01 3.349877e+01 -4.538915e-01 3.884324e+01 -4.675359e-01 4.504039e+01 -5.182642e-01 5.222624e+01 -6.086473e-01 6.055853e+01 -7.389039e-01 7.022019e+01 -9.071082e-01 8.142328e+01 -1.109675e+00 9.441375e+01 -1.341978e+00 1.094767e+02 -1.598965e+00 1.269429e+02 -1.875654e+00 1.471957e+02 -2.167486e+00 1.706797e+02 -2.470512e+00 1.979103e+02 -2.781477e+00 2.294854e+02 -3.097803e+00 2.660980e+02 -3.417531e+00 3.085519e+02 -3.739242e+00 3.577791e+02 -4.061963e+00 4.148600e+02 -4.385089e+00 4.810477e+02 -4.708303e+00 — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#2?email_source=notifications&email_token=ABG4UKUFXFW34272ZIPMHMLQJ4RK7A5CNFSM4IW6HL62YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HLP2X3Q>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABG4UKVTAORLNUP2DSCYFDLQJ4RK7ANCNFSM4IW6HL6Q> .

[xiaowu759]

It really works. Thank you very much!