let there be only one frame of indicies?
smoh opened this issue · comments
@adrn Right now, you cutoff stacked tgas catalog with a S/N cut in line 39 of generate-par-sample.py, and indicies generated from this file is for that particular S/N cut applied tgas, and we need to give compute-likelihood-ratio.py that same S/N. I think this is quite prone to mistakes... and it's better to only refer to indicies in the entire tgas catalog. This may be an issue when we pass info to @timothydmorton also. Let me know what you think.
@smoh You're absolutely right, I just couldn't figure out an easy way to do that without doing way more calculations than I needed.
My suggested changes in one-indicies branch -- I think that should get it done.
I didn't merge it because there's something else going on.
In master or one-indicies, I get the following for either random or nearest neighbor pairs.
projects/gaia-wide-binaries [master●] » python scripts/compute-likelihood-ratio.py -v --tgas-file ~/data/gaia/tgas_source/stacked_tgas.fits --pairs-file snr8_n32_dv4.fits --snr-cut 8
DEBUG: Using pool: <class 'schwimmbad.serial.SerialPool'> [__main__]
/Users/semyeong/anaconda2/lib/python2.7/site-packages/numpy/core/numeric.py:301: FutureWarning: in the future, full(3, 0) will return an array of dtype('int64')
format(shape, fill_value, array(fill_value).dtype), FutureWarning)
y [-22.77243541 -2.92362517 0. ]
M [[-0.12409192 -0.99227073 0. ]
[ 0.98235497 -0.12285187 0.14101818]
[-0.13992821 0.01749922 0.99000701]]
Cinv [[ 0.61678859 -0.03455493 0. ]
[-0.03455493 0.44979569 0. ]
[ 0. 0. 0. ]] [ 0. 0.44292787 0.6236564 ]
Vinv [[ 0. 0. 0.]
[ 0. 0. 0.]
[ 0. 0. 0.]] [ 0. 0. 0.]
Ainv [[ 0.45198487 0.05481973 0.06291484]
[ 0.05481973 0.60565471 -0.00295721]
[ 0.06291484 -0.00295721 0.00894469]] [ 3.67544536e-17 4.42927871e-01 6.23656402e-01]
Traceback (most recent call last):
File "scripts/compute-likelihood-ratio.py", line 151, in <module>
seed=args.seed, overwrite=args.overwrite)
File "scripts/compute-likelihood-ratio.py", line 89, in main
pool.map(worker, all_pairs, callback=worker.callback)
File "build/bdist.macosx-10.6-x86_64/egg/schwimmbad/serial.py", line 33, in map
File "scripts/compute-likelihood-ratio.py", line 33, in __call__
return self.work(i, star1, star2, v_scatter)
File "scripts/compute-likelihood-ratio.py", line 28, in work
n_dist_samples=self.n_distance_samples)
File "/Users/semyeong/projects/gaia-wide-binaries/gwb/fml.py", line 85, in ln_H2_FML
for d1,d2 in zip(dist1, dist2)])
File "/Users/semyeong/projects/gaia-wide-binaries/gwb/likelihood.py", line 142, in ln_H2_marg_v_likelihood
return (ln_Q(d1, data1, Vinv, v_scatter) + ln_Q(d2, data2, Vinv, v_scatter))
File "/Users/semyeong/projects/gaia-wide-binaries/gwb/likelihood.py", line 139, in ln_Q
return _marg_likelihood_helper(d, data, Vinv, v_scatter)
File "/Users/semyeong/projects/gaia-wide-binaries/gwb/likelihood.py", line 130, in _marg_likelihood_helper
assert sgn > 0
AssertionError
nvm..Vinv is zero because integer division happend in line 50 I think.
will consider this done. guess you'll probably add from __future__ import division.
Unless you already merged, can you file a PR from your branch to master so we can review there?
sorry, it's merged, and pushed.