notebook.community

Edit and run

I did this a bit flippantly before, but I want to fomalize the process by which we estimate the uncertainty on emulator predictions.



In [52]:

    
from pearce.emulator import NashvilleHot
from GPy.kern import *
import numpy as np
from os import path



In [53]:

    
import matplotlib
#matplotlib.use('Agg')
from matplotlib import pyplot as plt
%matplotlib inline
import seaborn as sns
sns.set()



In [54]:

    
'/scratch/users/swmclau2/xi_gg'









    Out[54]:





'/scratch/users/swmclau2/xi_gg'



In [69]:

    
training_file = '/scratch/users/swmclau2/xi_gg_zheng07_v5/PearceXiggCosmo.hdf5'
test_file = '/scratch/users/swmclau2/xi_gg_zheng07_cosmo_test_v4/PearceXiggCosmoTest.hdf5'

#xi gm training_file = '/scratch/users/swmclau2/xi_gm_cosmo/PearceRedMagicXiGMCosmoFixedNd.hdf5' test_file = '/scratch/users/swmclau2/xi_gm_cosmo_test2/PearceRedMagicXiGMCosmoFixedNdTest.hdf5'



In [70]:

    
em_method = 'gp'



In [71]:

    
fixed_params = {'z':0.0}

hyperparams = {'kernel': (Linear(input_dim=7, ARD=True) + RBF(input_dim=7, ARD=True)+Bias(input_dim=7), RBF(input_dim=4, ARD=True)+Bias(input_dim=4) ), \ 'optimize': True}



In [72]:

    
emu = NashvilleHot(training_file, hyperparams={},fixed_params = fixed_params, downsample_factor = 0.2)



In [73]:

    
pred_y, data_y = emu.goodness_of_fit(test_file, downsample_factor= 0.05, statistic = None)



In [74]:

    
data_y.shape









    Out[74]:





(18, 7000)



In [75]:

    
i = 3000
plt.plot(emu.scale_bin_centers, (data_y[:,i]-pred_y[:,i])/(data_y[:,i]))
#plt.plot(emu.scale_bin_centers, pred_y[:,i])
plt.xscale('log')
plt.show()



In [76]:

    
plt.plot( ((pred_y - data_y)/np.abs(data_y)).mean(axis =1) )
plt.ylim([0.0, 0.2])









    Out[76]:





(0.0, 0.2)



In [77]:

    
plt.plot( (np.abs(10**pred_y - 10**data_y)/(10**data_y)).mean(axis =1) )
plt.yscale('log')



In [27]:

    
emu.get_param_names()









    Out[27]:





['ombh2',
 'omch2',
 'w0',
 'ns',
 'ln10As',
 'H0',
 'Neff',
 'logM1',
 'logM0',
 'sigma_logM',
 'alpha',
 'conc_gal_bias']



In [ ]: