California-Planet-Search / rvsearch

RV Planet Search Pipeline Based on RadVel
http://california-planet-search.github.io/rvsearch/
MIT License
10 stars 5 forks source link

Error returned #176

Closed zhexingli closed 3 years ago

zhexingli commented 4 years ago

Hello, I'm getting an error below after running one rvsearch for hours:

Traceback (most recent call last): File "", line 1, in File "/Users/ZhexingLi/anaconda/lib/python3.6/site-packages/spyder/utils/site/sitecustomize.py", line 880, in runfile execfile(filename, namespace) File "/Users/ZhexingLi/anaconda/lib/python3.6/site-packages/spyder/utils/site/sitecustomize.py", line 102, in execfile exec(compile(f.read(), filename, 'exec'), namespace) File "/Users/ZhexingLi/Desktop/UCR/01-RESEARCH/PROJECTS/AAT/Targets/07-HD134330/HD134330Search.py", line 25, in searcher.run_search() File "/Users/ZhexingLi/anaconda/lib/python3.6/site-packages/rvsearch/search.py", line 529, in run_search save=False, ensembles=nensembles) File "/Users/ZhexingLi/anaconda/lib/python3.6/site-packages/radvel/mcmc.py", line 379, in mcmc minsteps=minsteps, minpercent=minpercent) File "/Users/ZhexingLi/anaconda/lib/python3.6/site-packages/radvel/mcmc.py", line 171, in convergence_check _status_message_CLI(statevars) File "/Users/ZhexingLi/anaconda/lib/python3.6/site-packages/radvel/mcmc.py", line 82, in _status_message_CLI statevars.screen = curses.initscr() File "/Users/ZhexingLi/anaconda/lib/python3.6/curses/init.py", line 30, in initscr fd=_sys.stdout.fileno()) _curses.error: setupterm: could not find terminal

Here's what my code looks like: path = '/Users/ZhexingLi/Desktop/UCR/01-RESEARCH/PROJECTS/WFIRST/TargetFit/08-HD142-c/data.txt' savepath = '/Users/ZhexingLi/Desktop/UCR/01-RESEARCH/PROJECTS/WFIRST/TargetFit/08-HD142-c/' data = pd.read_csv(path,header=None,skiprows=4,delim_whitespace=True,names=('time','mnvel','errvel','tel')) data['tel'] = 'AAT' searcher = search.Search(data, starname='142', min_per=1,workers=4, mcmc=True, \ verbose=True,fap=0.001,mstar=[1.232,0.22]) searcher.run_search() plotter = plots.PeriodModelPlot(searcher, saveplot=savepath+'PeriodSearch.png') plotter.plot_summary()

What's the cause of this? Are there any lines that have to be run from the command line rather than in the script?

Thanks for looking into this.

zhexingli commented 4 years ago

Same thing happened again... I think it happens every time when there's a signal above the 0.1% FAP level. Not sure why. Individual periodograms are produced and saved, rv fit are produced and saved. I tried to comment out the last two lines but still getting the same error message. It appears to be an issue related with run_search() function?

Is this function a command line only command? I'm running using a script. Is it why in the documentation it says its buggy in Jupyter notebook? Any way we can do this in the script?

I'm running a blind search so I don't have any pre-existing radvel files for my objects.

What else does this function produce? What am I missing?

zhexingli commented 4 years ago

Does thie error change the result of the periodogram or the rv fit at all? Is it safe to ignore this error message for now and still use rvsearch safely running from script?

leerosenthalj commented 4 years ago

I have never seen this error, it looks like it's mcmc-related and more about internal software than rvsearch. I would check your emcee installation, or even your bash setup; 'could not find terminal' sounds related to bash.

zhexingli commented 4 years ago

I've never touched my bash setup and I've never seen this error before too. What's the version of your curses package in python? Mine is 2.2, not sure if it's too old.

Radvel uses emcee as well right? Radvel doesn't seem to have any problem with emcee at all, and multiprocessing.

zhexingli commented 4 years ago

I can confirm, once mcmc is turned off, the error goes away. Weird... my radvel runs mcmc just fine