Lightrun
question-mark
Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

RuntimeError: Random walk sampling appears to be stuck!

See original GitHub issue

Hi,

I’m attempting to get dynesty to converge “nicely”. My model has 19 parameters and I run it with:

ndim = 19
dsampler = dynesty.DynamicNestedSampler(ln_like, prior_transform, ndim = ndim,bound='single',sample='rwalk',walks=50)

dsampler.run_nested(dlogz_init=1e-10, nlive_init=4000, nlive_batch=4000,wt_kwargs={'pfrac': 0.95})

Is setting dlogz_init=1e-10 an unrealistic value?

I’m not too sure how to get dynesty to always converge e.g. for mcmc I’d just increase the walkers and steps but I’ve found for dynesty that I need to alter the scale of the log likelihood function - where logl = non-normalised chi2 - otherwise the code runs for a relatively minimal time and the convergence isn’t great. If I leave the logl as just the non-normalised chi2 however, it prints the Runtime error. Any tips would be greatly appreciated.

Here’s the full print out (since the code says it may be useful)

RuntimeError: Random walk sampling appears to be stuck! Some useful output quantities:
u: [0.61211578 0.29883424 0.30418985 0.26553322 0.77596979 0.12880671
 0.53166602 0.55272972 0.59784004 0.47026351 0.49768702 0.46909061
 0.0492889  0.6416871  0.47095095 0.39649497 0.56340283 0.58003664
 0.43703319]
drhat: [ 0.51942188  0.20292128 -0.25131422 -0.0462435  -0.04087769 -0.03899258
 -0.28354396 -0.2750504  -0.45671211  0.22560593  0.04020417 -0.13088042
  0.05010286  0.10831015  0.06636851 -0.0941852   0.0129034   0.33036297
 -0.22233308]
dr: [ 0.50421269  0.19697954 -0.24395549 -0.04488944 -0.03968075 -0.03785084
 -0.27524151 -0.26699665 -0.44333913  0.21899996  0.03902695 -0.12704811
  0.0486358   0.10513872  0.06442517 -0.09142736  0.01252557  0.32068962
 -0.21582295]
du: [ 6.34458514e-03  1.41546553e-04 -3.43680742e-04 -2.97290180e-04
 -2.27000337e-03 -1.17278180e-03 -1.36977299e-03 -7.22953104e-05
 -2.04123144e-04  4.20805157e-04 -4.65647849e-05 -1.46280561e-04
 -2.21868830e-03  1.99202809e-03 -7.68809013e-04  1.79377016e-03
  8.00802805e-05  1.15701526e-03 -8.07340036e-04]
u_prop: [0.61211578 0.29883424 0.30418985 0.26553322 0.77596979 0.12880671
 0.53166602 0.55272972 0.59784004 0.47026351 0.49768702 0.46909061
 0.0492889  0.6416871  0.47095095 0.39649497 0.56340283 0.58003664
 0.43703319]
loglstar: -989554.5
logl_prop: -989554.8125
axes: [[ 1.25831524e-02  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [-9.59494844e-06  7.43145437e-04  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 3.71301087e-05 -1.15653354e-03  5.51693936e-04  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 6.94499246e-06 -1.69150355e-04  8.73413246e-04  1.21183982e-03
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 2.73957906e-04 -8.60692885e-03  2.74503412e-03 -8.04795262e-06
   1.09480371e-03  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 1.24898412e-04 -5.24219719e-03  7.02620596e-04 -3.72378230e-04
   5.08216528e-04  7.47482506e-04  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [-3.19867398e-04 -1.66256241e-03 -4.87165649e-04  6.37138555e-04
  -2.21943007e-03  4.31434478e-04  3.78934642e-03  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 6.99011461e-05  7.50939116e-05 -3.75765342e-04 -5.11191945e-04
   8.43199529e-05 -4.94147094e-05 -3.04046425e-04  1.19537144e-03
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 2.09550450e-05  3.67508706e-05 -3.12002971e-05 -1.24150632e-04
   1.43859932e-04  8.81744894e-05 -2.82908663e-04  1.03809279e-04
   6.23040387e-04  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 3.05516274e-05  2.70754057e-04  1.29882321e-04 -1.00638958e-04
   2.85437772e-04  7.13926851e-07 -5.20945605e-04 -1.08674654e-04
  -1.27351404e-04  7.38481547e-04  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [-1.38433571e-04 -3.04582253e-05 -1.36178139e-05 -2.89814475e-05
   7.80309440e-05 -3.20340557e-05  3.34764725e-05 -1.31236383e-05
  -7.06886466e-06 -3.81492237e-05  9.58990041e-04  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [-1.92016931e-05 -7.64026977e-06  6.12720171e-05 -1.42674137e-04
  -2.05555985e-05  3.68160631e-05 -7.12946160e-06  3.92718877e-05
   2.26589176e-05  3.33146619e-06  3.96333729e-04  9.72871403e-04
   0.00000000e+00  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [-4.41093062e-03  1.89035515e-05  4.67148659e-06 -3.70515645e-06
  -1.84964955e-05 -3.39352322e-06  1.38457554e-05  3.16455044e-06
   8.71002817e-06  8.37528244e-06 -1.22932324e-05  7.23592550e-08
   1.83369330e-04  0.00000000e+00  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [-3.35641371e-04  6.84401918e-03 -3.36150827e-03  4.37332517e-04
  -8.62379949e-04  1.61894799e-04  3.51631144e-04 -6.95863825e-05
   1.58963842e-04  1.01999644e-04  6.54033393e-05  3.89800472e-05
   1.73473518e-05  1.07010444e-03  0.00000000e+00  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 1.77472187e-04 -4.75235561e-03 -3.20157828e-04 -3.03513420e-04
   1.04439265e-03  1.67462704e-04 -9.77002001e-05 -1.03067467e-05
   4.51667620e-05  1.45626808e-04  2.11614695e-05 -4.69488571e-05
  -6.59153625e-05 -5.98547171e-04  8.02648703e-04  0.00000000e+00
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [ 9.21585938e-05  1.20520033e-03 -6.63482858e-03 -7.46695507e-03
   1.98472258e-04  5.08495839e-05  3.68956297e-04  3.08721385e-04
  -2.93650482e-04 -3.06530234e-04  5.83505785e-05  1.67152074e-05
   8.41808224e-05 -2.35639096e-04 -8.68709842e-04  2.58861780e-03
   0.00000000e+00  0.00000000e+00  0.00000000e+00]
 [-2.35606757e-04 -1.01496058e-03  2.06209330e-04  3.83653812e-04
  -5.04916312e-04 -2.67252253e-03 -5.29116257e-04 -1.79550685e-04
  -2.60915858e-04 -3.75345682e-04 -1.27468841e-04 -1.57467145e-04
   7.57737881e-05  1.41913802e-04 -1.70889529e-03 -1.76620976e-03
   2.64012072e-03  0.00000000e+00  0.00000000e+00]
 [ 2.60843480e-04 -4.84038713e-05  7.51183672e-04  3.27388573e-04
   5.05505263e-04 -5.59759483e-04 -1.10650854e-03 -6.76308608e-05
   3.86106624e-05 -3.05519116e-04 -9.08987536e-05 -6.56174791e-05
  -1.32180212e-04 -8.39534513e-05 -1.36644947e-04 -1.06686592e-03
   1.91169322e-04  2.84573877e-03  0.00000000e+00]
 [-1.11306400e-04 -2.84578527e-04 -3.70878506e-04 -5.95121814e-05
   7.12390082e-04  8.54350560e-04  1.62642821e-03 -1.43116178e-04
   2.63919121e-04  3.27381262e-04  7.70220025e-05  1.31537510e-04
   1.39954314e-04  1.00476004e-03 -2.31798803e-04  4.33121317e-04
   1.87592818e-04 -1.22857147e-05  1.46214361e-03]]
scale: 6.699929041550063e-13.

Issue Analytics

  • State:closed
  • Created 2 years ago
  • Comments:9 (1 by maintainers)

github_iconTop GitHub Comments

1reaction
segasaicommented, Aug 13, 2021

Okay, I’ve read your previous posts in detail And it’s seems your likelihood f-n needs to be something like sum(-0.5*((data-model)^2/err^2 -ln(err))
where err is an extra-parameter for the noise of your data. (that is making an assumption that all your measurements are independent and share the same uncertainty.

0reactions
bjnorfolkcommented, Aug 16, 2021

@segasai thanks for the suggestion, I went with something similar and the code now nicely converges. Thanks!

Read more comments on GitHub >

github_iconTop Results From Across the Web

Sampling gets stuck after repeated instances of "UserWarning ...
Sampling gets stuck after repeated instances of "UserWarning: Random walk proposals appear to be extremely inefficient." #160.
Read more >
Random Walk(Python Crash Course) stuck in what seems to ...
When I run the code, the command prompt gets stuck in a "processing" mode" which I cannot leave unless I close the window,...
Read more >
How do I run Stable Diffusion and sharing FAQs - Reddit
I installed git manually but now when I do env create -f environment.yaml it seems to get stuck when installing pip dependencies. No...
Read more >
emcee - Release 3.0.2
During sampling, at each step, a move is randomly selected from the mixture and used as the proposal. The default move is still...
Read more >
Random walk on a cube | Physics Tom
This is easy to see if examine k=1,2,3,4 and 5. which is obvious - there is one path from 1 → 2, one...
Read more >

github_iconTop Related Medium Post

No results found

github_iconTop Related StackOverflow Question

No results found

github_iconTroubleshoot Live Code

Lightrun enables developers to add logs, metrics and snapshots to live code - no restarts or redeploys required.
Start Free

github_iconTop Related Reddit Thread

No results found

github_iconTop Related Hackernoon Post

No results found

github_iconTop Related Tweet

No results found

github_iconTop Related Dev.to Post

No results found

github_iconTop Related Hashnode Post

No results found
Previous page

Failures in tests

Next page

Cannot compute a bounding ellipsoid to a single point if `pointvol` is not specified