Richard Elkins

There was an incomplete work-around here previously. Withdrawn because it was missing the cmake build steps.

@aSmig Thanks, you are spot on. Doing too many things at the same time. The Python scripts are updated but the cmake steps need to be completed. Sorry for the...

Yes, the above program turned my PC into a swapping machine. I tried some variations using `del frame` at the end of the loop instead of `pass`. Garbage collection caught...

Note that ``` from lightkurve import KeplerTargetPixelFile TPF_NAME = "https://archive.stsci.edu/pub/kepler/target_pixel_files/0069/006922244/kplr006922244-2010078095331_lpd-targ.fits.gz" tpf = KeplerTargetPixelFile(TPF_NAME) print("Loaded tpf, # of cadences=", len(tpf.cadenceno)) for ndx, frame in enumerate(tpf.flux): pass print("I finished!") ``` runs just...

I re-installed lightkurve from github. My experience with ```python -m memory_profiler tpf_leak.py``` was as follows: ``` numpy 1.19.1 astropy 4.1rc1 lightkurve 2.0a2.dev (18104, 21, 11) 4182024 Filename: tpf_leak.py Line #...

Haven't been around lightkurve much lately because of commitments to cryptography projects. But, memory leaks are critical issues. When I have time, I'll look at the tests some more.

@drsamadi I modified your source code to show the iterations: ``` import numpy as np import lightkurve as lk nloops = 10000 TICID = np.ones(nloops,dtype=np.int64)*85148268 counter = 0 for ticid...

@drsamadi You still have not described your O/S type/version and hardware except for having 16GB RAM. I'd guess that you are using some version of Windows. True? My machine: -...

@drsamadi Okay then open a NEW issue since this one is closed based on an older lightkurve: * Show characteristics like you just did (copy/paste/edit). * Explain 2.4.0 vs 2.4.1.

@drsamadi With 2.4.1, I can now get into memory trouble. Spyder at loop 41: "Kernel died, restarting" @barentsen lightkurve-2.4.1.dev0, should we be avoiding this? I got it by this method:...