beckdaniel

import numpy as np
import GPy as gp

X = np.random.uniform(-3.,3.,(2000,17))
Y = np.sin(X) + np.random.randn(2000,17)*0.05
kernel = gp.kern.rbf(D=17, variance=1., lengthscale=1.)
m = gp.models.GP_regression(X,Y,kernel,normalize_X=True,normalize_Y=True)
m.constrain_positive('')
m.optimize()

beckdaniel

danielb@elgon:~/Research/Datasets/QE$ head -1 eamt11/fr-en/feats_17.tsv 
13.0	15.0	4.38461538462	-31.6027	-34.8847	0.8	310.461538462	1185.38461538	0.153846153846	15632.0	0.166666666667	16934.6666667	0.181818181818	18474.1818182	1.0	0.0769230769231	0.0666666666667
danielb@elgon:~/Research/Datasets/QE$ head -1 wptp-12_daniel/qe_17 
19.0	17.0	4.4210525	-43.948	-39.2668	1.0	144.0	0.04820285	0.0	0.7894737	0.0	0.5	0.0	0.23529412	0.8947368	2.0	2.0

beckdaniel

Python 2.7.10 |Anaconda 2.4.0 (64-bit)| (default, Oct 19 2015, 18:04:42) 
Type "copyright", "credits" or "license" for more information.

IPython 4.0.0 -- An enhanced Interactive Python.
?         -> Introduction and overview of IPython's features.
%quickref -> Quick reference.
help      -> Python's own help system.
object?   -> Details about 'object', use 'object??' for extra details.

In [1]: import numpy as np

beckdaniel

import numpy as np
import GPy
from sklearn.svm import SVR
import sys
import memory_profiler as mprof

model = sys.argv[1]

feats = np.random.random(size=(100,10))
labels = np.random.random(size=(100,1))

beckdaniel

import GPy
import GPflow
import numpy as np

rng = np.random.RandomState(0)
X = rng.randn(100,2)
Y = rng.randn(100,1)
Xtest = rng.randn(10,2)

# GPy

beckdaniel

import GPy
import GPflow
import numpy as np

rng = np.random.RandomState(0)
X = rng.randn(1000,2)
Y = rng.randn(1000,1)
Xtest = rng.randn(10,2)

# GPy