In [1]:

    

import numpy as np
import pandas as pd
from ggplot import *
import simple_access
import statsmodels.api as sm
from scipy import stats
%matplotlib inline


    

In [29]:

    

sel = [ 'ptNF>5.0', 'ptNF<200.0','qzpartOF>-2.25',
       'qzpartOF<2.25','prpart1OF>-1.25', 'prpart1OF<1.25',
       'prpart2OF>-1.25', 'prpart2OF<1.25', 'qrpart1OF>-1.25',
       'qrpart1OF<1.25','qrpart2OF>-1.25', 'qrpart2OF<1.25']
sel2 = ['cGoodEv_v53==1.0', 'cRandom_133==0.0']
s1 = ' and '.join(sel[0:4]+sel2)


    

In [4]:

    

rec = simple_access.chainer(izip=7,
                            rrqs=['pzpartOF', 'ptNF', 'prpart1OF', 'prpart2OF', 'qzpartOF',
                                  'qrpart1OF','qrpart2OF'],
                            cuts=['cGoodEv_v53', 'cGoodDataPeriod_133', 'cGoodDataPeriod_v53',
                                  'cGoodPulse_v53', 'cGoodPulseQ_v53', 'cGoodPulseP_v53',
                                  'cPChiSq_v53', 'cPstd_v53', 'cGoodPStartTime_v53', 'cLFnoise1_v53',
                                  'cGlitch1_v53', 'cQChiSq_v53', 'cQstd_v53', 'cGoodBiasTime_133',
                                  'cBadSeries_133', 'cTrigBurst_133', 'cFinalPhononSettings_133',
                                  'cGoodDCOffset_v53', 'cGoodBaseTemp_v53', 'cQhighnoise_v53', 'cSquarePulse_v53',
                                  'cBadLED_v53', 'cBadOFRes_v53'],
                            eventcuts=['cRandom_133', 'cBadSeries_133', 'cTrigBurst_133', 'cErrMask_133', 'cGlitch_133'],
                            dtype='cf',
                            load_cut_to_ram=True)


    

    




Load time:  42.5518240929 s

In [ ]:

    

data = df[['ptNF','pzpartOF']]
kern = sm.nonparametric.KDEMultivariate(data = data.values, var_type='cc', bw='normal_reference')


    

In [24]:

    

p = ggplot(aes(x='ptNF', y='qzpartOF'), data=rec.query(s1)) + geom_point(alpha=0.5)
p


    

    













    
Out[24]:





<ggplot: (14069393)>

In [30]:

    

p_color = ggplot(aes(x='ptNF', y='qzpartOF', color = 'qzpartOF'), data=rec.query(s1)) + geom_point(alpha=0.5) + scale_colour_gradient2(low='blueviolet', high='darkorange')


    

In [31]:

    

p_color


    

    













    
Out[31]:





<ggplot: (11721529)>

In [30]:

    

df = rec.query(s1)
H = np.histogram2d(df.ptNF.values, df.qzpartOF.values, bins=30)
h1 = pd.DataFrame(H[0], index=H[1][1:], columns=H[2][1:])
h2 = pd.DataFrame(H[0].T, index=H[2][1:], columns=H[1][1:])
h1['ptNF'] = h1.index
h2['qzpartOF'] = h2.index
h1_melt=pd.melt(h1, id_vars=['ptNF'])
h2_melt = pd.melt(h2, id_vars=['qzpartOF'])


    

In [31]:

    

p1 = ggplot(aes(x='ptNF', y='value'), data=h1_melt) + geom_step() + facet_wrap("variable")
p2 = ggplot(aes(x='qzpartOF', y='value'), data=h2_melt) + geom_step() + facet_wrap("variable")


    

In [32]:

    

p1


    

    













    
Out[32]:





<ggplot: (19111509)>

In [33]:

    

p2


    

    













    
Out[33]:





<ggplot: (14134149)>

In [11]:

    




    

In [12]:

    

data.values.shape


    

    
Out[12]:





(22663, 2)

In [13]:

    




    

In [16]:

    

kern.pdf(data.ix[1])


    

    
Out[16]:





array(0.003506839988981963)

In [35]:

    

reload(simple_access)


    

    
Out[35]:





<module 'simple_access' from 'simple_access.py'>

In [ ]: