Roche::ingress_egress raises linmin error about cpp-roche HOT 5 OPEN

trmrsh commented on August 17, 2024

Roche::ingress_egress raises linmin error

from cpp-roche.

Comments (5)

trmrsh commented on August 17, 2024

Hi Stu,

yes, I am aware that it sometimes fails. Is this a problem for you for a
real computation or just with the py2 --> 3 issue?

Its useful to have the exact numbers here, and I will try to take a look at
it to see if I can fix it.

tom

On 12 October 2015 at 14:17, StuartLittlefair [email protected]
wrote:

When calling ingress_egress with q=0.0578808 and i=85.379 to see if a
point at 2.47587 -2.5694 0 is eclipsed by the donor star I get the
following error output

Roche::linmin failed to bracket minimum, error 2

It's unclear how to deal with this.

—
Reply to this email directly or view it on GitHub
#2.

from cpp-roche.

StuartLittlefair commented on August 17, 2024

It's not a serious problem, but I've moved LFIT to use ingress_egress in an
attempt to speed things up. Previously I used blink a lot. It now fails
occasionally.

I'm just assuming that when it does this the point is not eclipsed for now
and swallowing the error...

On 13 October 2015 at 10:35, Tom [email protected] wrote:

Hi Stu,

yes, I am aware that it sometimes fails. Is this a problem for you for a
real computation or just with the py2 --> 3 issue?

Its useful to have the exact numbers here, and I will try to take a look at
it to see if I can fix it.

tom

On 12 October 2015 at 14:17, StuartLittlefair [email protected]
wrote:

When calling ingress_egress with q=0.0578808 and i=85.379 to see if a
point at 2.47587 -2.5694 0 is eclipsed by the donor star I get the
following error output

Roche::linmin failed to bracket minimum, error 2

It's unclear how to deal with this.

—
Reply to this email directly or view it on GitHub
#2.

—
Reply to this email directly or view it on GitHub
#2 (comment).

Stuart Littlefair

Dept. of Physics & Astronomy,
Univ. of Sheffield, Sheffield, S3 7RH.

email: [email protected]
phone: +44 114 2224525

from cpp-roche.

trmrsh commented on August 17, 2024

I run into it as well now and then but have always managed to escape it,
but I should see if I can fix it.

tom

On 13 October 2015 at 10:46, StuartLittlefair [email protected]
wrote:

It's not a serious problem, but I've moved LFIT to use ingress_egress in an
attempt to speed things up. Previously I used blink a lot. It now fails
occasionally.

I'm just assuming that when it does this the point is not eclipsed for now
and swallowing the error...

On 13 October 2015 at 10:35, Tom [email protected] wrote:

Hi Stu,

yes, I am aware that it sometimes fails. Is this a problem for you for a
real computation or just with the py2 --> 3 issue?

Its useful to have the exact numbers here, and I will try to take a look
at
it to see if I can fix it.

tom

On 12 October 2015 at 14:17, StuartLittlefair [email protected]
wrote:

When calling ingress_egress with q=0.0578808 and i=85.379 to see if a
point at 2.47587 -2.5694 0 is eclipsed by the donor star I get the
following error output

Roche::linmin failed to bracket minimum, error 2

It's unclear how to deal with this.

—
Reply to this email directly or view it on GitHub
#2.

—
Reply to this email directly or view it on GitHub
#2 (comment).

Stuart Littlefair

Dept. of Physics & Astronomy,
Univ. of Sheffield, Sheffield, S3 7RH.

email: [email protected]
phone: +44 114 2224525

—
Reply to this email directly or view it on GitHub
#2 (comment).

from cpp-roche.

wildjames commented on August 17, 2024

I think I'm experiencing a version of this issue. I get an error many times over an MCMC chain (the frequency seems to vary) that kicks back with this:

Diagnostics
xb,xc,fb,fc= 1,1,,-2.09253,-2.09253
q,ci,si,p = 0.0503799, 0.0777121, 0.996976, 2.41302 -2.59534 0
phi, lam, dphi, dlam = 0.672343, 2.94606, 0, -0.0312574
phi1,phi2,lam1,lam2 = 0.668958, 0.672343, 2.9148, 2.97746
pref, acc = -2.20445, 0.000177465
0 0 -2.09122 -0.0028331
1 0.02 -2.09127 -0.00277074
2 0.04 -2.09133 -0.00270842
3 0.06 -2.09138 -0.00264612
4 0.08 -2.09144 -0.00258387
5 0.1 -2.09149 -0.00252166
6 0.12 -2.09154 -0.0024595
7 0.14 -2.09158 -0.00239739
8 0.16 -2.09163 -0.00233534
9 0.18 -2.09168 -0.00227335
10 0.2 -2.09172 -0.00221143
11 0.22 -2.09177 -0.00214959
12 0.24 -2.09181 -0.00208782
13 0.26 -2.09185 -0.00202613
14 0.28 -2.09189 -0.00196453
15 0.3 -2.09193 -0.00190303
16 0.32 -2.09197 -0.00184162
17 0.34 -2.092 -0.00178031
18 0.36 -2.09204 -0.00171911
19 0.38 -2.09207 -0.00165802
20 0.4 -2.0921 -0.00159704
21 0.42 -2.09213 -0.00153619
22 0.44 -2.09216 -0.00147546
23 0.46 -2.09219 -0.00141486
24 0.48 -2.09222 -0.0013544
25 0.5 -2.09225 -0.00129407
26 0.52 -2.09227 -0.00123389
27 0.54 -2.0923 -0.00117386
28 0.56 -2.09232 -0.00111399
29 0.58 -2.09234 -0.00105427
30 0.6 -2.09236 -0.000994709
31 0.62 -2.09238 -0.00093532
32 0.64 -2.0924 -0.000876101
33 0.66 -2.09242 -0.000817059
34 0.68 -2.09243 -0.000758197
35 0.7 -2.09245 -0.00069952
36 0.72 -2.09246 -0.000641032
37 0.74 -2.09247 -0.000582738
38 0.76 -2.09248 -0.000524642
39 0.78 -2.09249 -0.000466748
40 0.8 -2.0925 -0.000409061
41 0.82 -2.09251 -0.000351584
42 0.84 -2.09252 -0.000294323
43 0.86 -2.09252 -0.000237281
44 0.88 -2.09253 -0.000180462
45 0.9 -2.09253 -0.00012387
46 0.92 -2.09253 -6.75097e-05
47 0.94 -2.09253 -1.1385e-05
48 0.96 -2.09253 4.45002e-05
49 0.98 -2.09253 0.000100142
50 1 -2.09253 0.000155536
ValueError('setup grid failed with linmin error')
Error: setup grid failed with linmin error; parlist: [0.09021958287754558, 0.026674351652483153, 0.013744000187760196, 0.00025009950737006327, 0.05037991581497417, 0.04429157722643255, 0.4095789506968387, 0.44649996165818984, 0.017277431457750726, 0.013617261084380886, 127.87002855562842, 0.43196900033098856, 1.7821816566694784, -5.177135543962317e-05, 2.521564860791719, 0.34902572616817606, 74.43299412020373, -25.668439873507573]

I can provide a decent number of parameter lists that produce this if you like, but for now I'll just post two. Here's another message:

Diagnostics
xb,xc,fb,fc= 1,1,,-2.11025,-2.11025
q,ci,si,p = 0.266106, 0.221297, 0.975206, 1.02412 1.65906 0
phi, lam, dphi, dlam = 0.250271, 1.61798, 0, -0.0206471
phi1,phi2,lam1,lam2 = 0.250271, 0.254357, 1.59733, 1.63887
pref, acc = -2.38773, 0.000144493
0 0 -2.10944 -0.0016429
1 0.02 -2.10947 -0.00160972
2 0.04 -2.1095 -0.00157654
3 0.06 -2.10953 -0.00154336
4 0.08 -2.10956 -0.00151018
5 0.1 -2.10959 -0.001477
6 0.12 -2.10962 -0.00144382
7 0.14 -2.10965 -0.00141065
8 0.16 -2.10968 -0.00137748
9 0.18 -2.10971 -0.00134431
10 0.2 -2.10973 -0.00131115
11 0.22 -2.10976 -0.00127799
12 0.24 -2.10978 -0.00124484
13 0.26 -2.10981 -0.00121169
14 0.28 -2.10983 -0.00117854
15 0.3 -2.10985 -0.0011454
16 0.32 -2.10988 -0.00111227
17 0.34 -2.1099 -0.00107915
18 0.36 -2.10992 -0.00104603
19 0.38 -2.10994 -0.00101292
20 0.4 -2.10996 -0.000979813
21 0.42 -2.10998 -0.000946718
22 0.44 -2.11 -0.000913632
23 0.46 -2.11002 -0.000880555
24 0.48 -2.11003 -0.000847487
25 0.5 -2.11005 -0.00081443
26 0.52 -2.11007 -0.000781383
27 0.54 -2.11008 -0.000748347
28 0.56 -2.1101 -0.000715322
29 0.58 -2.11011 -0.000682308
30 0.6 -2.11012 -0.000649307
31 0.62 -2.11014 -0.000616318
32 0.64 -2.11015 -0.000583341
33 0.66 -2.11016 -0.000550378
34 0.68 -2.11017 -0.000517429
35 0.7 -2.11018 -0.000484494
36 0.72 -2.11019 -0.000451573
37 0.74 -2.1102 -0.000418667
38 0.76 -2.11021 -0.000385776
39 0.78 -2.11021 -0.000352902
40 0.8 -2.11022 -0.000320043
41 0.82 -2.11023 -0.0002872
42 0.84 -2.11023 -0.000254375
43 0.86 -2.11024 -0.000221567
44 0.88 -2.11024 -0.000188776
45 0.9 -2.11024 -0.000156004
46 0.92 -2.11025 -0.00012325
47 0.94 -2.11025 -9.05157e-05
48 0.96 -2.11025 -5.78002e-05
49 0.98 -2.11025 -2.51044e-05
50 1 -2.11025 7.57122e-06
ValueError('setup grid failed with linmin error')
Error: setup grid failed with linmin error; parlist: [0.12006416024839853, 0.05492891216742712, 0.15431241965402165, 0.008651142508537202, 0.26610633216414287, 0.04194458923390031, 0.44616471570591865, 0.28385990968989805, 0.03525109079315824, 0.10064684725508243, 63.213504141460774, 0.01928129361370185, 0.08466085793302407, -0.0024301974175894026, 0.8051254579358218, 1.4497382465152464, 51.82644021750187, 65.35239013787086]

from cpp-roche.

trmrsh commented on August 17, 2024

Thanks James. The job its trying to do is locate a minimum in the Roche potential between two points, and presumably not finding one. It is a long running issue that needs some work; I won't have the time for some time to tackle it I'm afraid, just got too much on. I am sure it's fixable.

from cpp-roche.

Roche::ingress_egress raises linmin error about cpp-roche HOT 5 OPEN

Comments (5)

Related Issues (1)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent