mcmc error - Probability function returned NaN

[aberrico]

I'm tring to run mcmc using this command: radvel mcmc -s /path/file.py And I'm getting this error message: [======================== ] 1880000/4000000 (47.0%) steps complete; Running 2077.06 steps/s; Mean acceptance rate = 16.0%; Min Auto Factor = 17; Max Auto Relative-Change = 0.0337; Min Tz = 353.4; Max G-R = 1.090 Traceback (most recent call last): File "/home/adriana/anaconda3/bin/radvel", line 8, in sys.exit(main()) File "/home/adriana/anaconda3/lib/python3.7/site-packages/radvel/cli.py", line 194, in main args.func(args) File "/home/adriana/anaconda3/lib/python3.7/site-packages/radvel/driver.py", line 198, in mcmc savename=backend_loc, proceed=args.proceed, proceedname=backend_loc) File "/home/adriana/anaconda3/lib/python3.7/site-packages/radvel/mcmc.py", line 363, in mcmc for sample in sampler.sample(p1, store=True): File "/home/adriana/anaconda3/lib/python3.7/site-packages/emcee-3.0.2-py3.7.egg/emcee/ensemble.py", line 343, in sample state, accepted = move.propose(model, state) File "/home/adriana/anaconda3/lib/python3.7/site-packages/emcee-3.0.2-py3.7.egg/emcee/moves/red_blue.py", line 93, in propose new_log_probs, new_blobs = model.compute_log_prob_fn(q) File "/home/adriana/anaconda3/lib/python3.7/site-packages/emcee-3.0.2-py3.7.egg/emcee/ensemble.py", line 456, in compute_log_prob raise ValueError("Probability function returned NaN") ValueError: Probability function returned NaN

[bjfultn]

This usually happens when your fit is poorly constrained or your starting guesses for parameters are outside or near the boundaries of priors.