def test1(self):
R="""Bivariate Correlations
A B C
======================================================
A spearman 1 0.958 -0.924
Sig (2-tailed) . 9.699e-12 2.259e-09
N 21 21 21
------------------------------------------------------
B spearman 0.958 1 -0.890
Sig (2-tailed) 9.699e-12 . 0.000
N 21 21 21
------------------------------------------------------
C spearman -0.924 -0.890 1
Sig (2-tailed) 2.259e-09 0.000 .
N 21 21 21
Larzelere and Mulaik Significance Testing
Pair i Correlation P alpha/(k-i+1) Sig.
============================================================
A vs. B 1 0.958 9.699e-12 0.017 **
A vs. C 2 0.924 2.259e-09 0.025 **
B vs. C 3 0.890 6.850e-08 0.050 ** """
df=DataFrame()
df['A']=[24,61,59,46,43,44,52,43,58,67,62,57,71,49,54,43,53,57,49,56,33]
df['B']=[42.93472681237495, 78.87307334936268, 75.37292628918023, 65.49076317291956, 55.55965179772366, 56.777730638998236, 62.19451880792437, 54.73710611356715, 72.10021832823149, 85.94377749485642, 78.2087578930983, 72.01681829338037, 84.27889316830063, 60.20516982367225, 65.6276497088971, 62.36549856901088, 69.18772114281175, 67.00548667483324, 59.042687027269466, 71.99214593063917, 45.00831155783992]
df['C']=[-53.05540625388731, -96.33996451998567, -92.32465861908086, -70.90536432779966, -55.953777697739255, -74.12814626217357, -75.89188834814621, -64.24093256012688, -89.62208010083313, -87.41075066046812, -80.40932820298143, -77.99906284144805, -95.31607277596169, -61.672429800914486, -85.26088499198657, -63.4402296673869, -74.84950736563589, -85.00433219746624, -71.5901436929124, -76.43243666219388, -48.01082320924727]
cor=df.correlation(['A','B','C'],coefficient='spearman')
self.assertEqual(str(cor),R)
def test02(self):
df=DataFrame()
df.read_tbl('data/words~ageXcondition.csv')
D = repr(df.histogram('WORDS'))
R = "Histogram([('values', [4.0, 14.0, 17.0, 12.0, 15.0, 10.0, 9.0, 5.0, 6.0, 8.0]), \
('bin_edges', [3, 5.0, 7.0, 9.0, 11.0, 13.0, 15.0, 17.0, 19.0, 21.0, 23])], cname='WORDS')"
self.assertEqual(D, R)
def test01(self):
"""repr test"""
R = Descriptives([('count', 100.0),
('mean', 11.61),
('mode', 11.0),
('var', 26.947373737373752),
('stdev', 5.191085988246944),
('sem', 0.5191085988246944),
('rms', 12.707084638106414),
('min', 3.0),
('Q1', 7.0),
('median', 11.0),
('Q3', 15.5),
('max', 23.0),
('range', 20.0),
('95ci_lower', 10.592547146303598),
('95ci_upper', 12.6274528536964)],
cname='WORDS')
df = DataFrame()
df.read_tbl('data/words~ageXcondition.csv')
D = eval(repr(df.descriptives('WORDS')))
for k in D.keys():
self.failUnlessAlmostEqual(D[k],R[k])
def test_kn(self):
df = DataFrame()
df.read_tbl('data/example.csv')
y = [23]*len(df['X'])
df['X'] = y
self.assertEqual(df.keys(), ['CASE', 'TIME', 'CONDITION', 'X'])
def test2(self):
R = DataFrame([('SUBJECT', [1, 2]), ('TIMEOFDAY', [u'T1', u'T1']), ('COURSE', [u'C1', u'C2']), ('MODEL', [u'M1', u'M1']), ('ERROR', [10, 10])])
df=DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
df2 = df.where([('ERROR', '=', 10)])
self.assertEqual(repr(df2),repr(df2))
def test6(self):
R = DataFrame([('SUBJECT', [1, 1, 1, 1, 1, 1, 2, 2, 2, 3, 3, 3, 3, 3, 3]), ('TIMEOFDAY', [u'T1', u'T1', u'T1', u'T2', u'T2', u'T2', u'T2', u'T2', u'T2', u'T1', u'T1', u'T1', u'T2', u'T2', u'T2']), ('COURSE', [u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1', u'C1']), ('MODEL', [u'M1', u'M2', u'M3', u'M1', u'M2', u'M3', u'M1', u'M2', u'M3', u'M1', u'M2', u'M3', u'M1', u'M2', u'M3']), ('ERROR', [10, 8, 6, 5, 4, 3, 4, 3, 3, 8, 7, 4, 4, 1, 2])])
df=DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
df2 = df.where([('COURSE','=',['C1']),('TIMEOFDAY','in',["T1", "T2"])])
self.assertEqual(repr(df2),repr(R))
def test2(self):
R="""\
Chi-Square: Single Factor
SUMMARY
1 2 3 4 5
=====================================================
Observed 7 20 23 9 0
Expected 11.800 11.800 11.800 11.800 11.800
CHI-SQUARE TESTS
Value df P
============================================
Pearson Chi-Square 30.746 4 3.450e-06
Likelihood Ratio -- -- --
Observations 59
POST-HOC POWER
Measure
==============================
Effect size w 0.722
Non-centrality lambda 30.746
Critical Chi-Square 9.488
Power 0.998 """
df = DataFrame()
df.read_tbl('data/chi_test.csv')
X=df.chisquare1way('RESULT',{1:11.8 ,2:11.8 ,3:11.8 ,4:11.8 ,5:11.8})
self.assertEqual(str(X),R)
def test1(self):
R = {'d': [np.array([ 9, 8, 6, 8, 10, 4, 6, 5, 7, 7,
7, 9, 6, 6, 6, 11, 6, 3, 8, 7,
11, 13, 8, 6, 14, 11, 13, 13, 10, 11,
12, 11, 16, 11, 9, 23, 12, 10, 19, 11,
10, 19, 14, 5, 10, 11, 14, 15, 11, 11]),
np.array([ 8, 6, 4, 6, 7, 6, 5, 7, 9, 7,
10, 7, 8, 10, 4, 7, 10, 6, 7, 7,
14, 11, 18, 14, 13, 22, 17, 16, 12, 11,
20, 16, 16, 15, 18, 16, 20, 22, 14, 19,
21, 19, 17, 15, 22, 16, 22, 22, 18, 21])],
'fname': 'output\\box(WORDS~AGE).png',
'maintitle': 'WORDS by AGE',
'xlabels': [u'AGE = old', u'AGE = young']}
df=DataFrame()
df.TESTMODE=True
df.read_tbl('data/words~ageXcondition.csv')
D=df.box_plot('WORDS',['AGE'], output_dir='output')
self.assertEqual(D['fname'],R['fname'])
self.assertEqual(D['maintitle'],R['maintitle'])
self.assertEqual(D['xlabels'],R['xlabels'])
for d,r in zip(np.array(D['d']).flat,
np.array(R['d']).flat):
self.assertAlmostEqual(d,r)
def test0(self):
df=DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
pt = df.pivot('ERROR', ['TIMEOFDAY'],['COURSE'])
self.assertAlmostEqual(3.22222222222,pt[1,0],5)
def test2(self):
## Between-Subjects test
df=DataFrame()
fname='words~ageXcondition.csv'
df.read_tbl(fname)
aov=Anova()
aov.run(df,'WORDS',bfactors=['AGE','CONDITION'])
def test0(self):
R = {'d': [9.0, 8.0, 6.0, 8.0, 10.0, 4.0, 6.0, 5.0, 7.0, 7.0,
7.0, 9.0, 6.0, 6.0, 6.0, 11.0, 6.0, 3.0, 8.0, 7.0,
11.0, 13.0, 8.0, 6.0, 14.0, 11.0, 13.0, 13.0, 10.0,
11.0, 12.0, 11.0, 16.0, 11.0, 9.0, 23.0, 12.0, 10.0,
19.0, 11.0, 10.0, 19.0, 14.0, 5.0, 10.0, 11.0, 14.0,
15.0, 11.0, 11.0, 8.0, 6.0, 4.0, 6.0, 7.0, 6.0, 5.0,
7.0, 9.0, 7.0, 10.0, 7.0, 8.0, 10.0, 4.0, 7.0, 10.0,
6.0, 7.0, 7.0, 14.0, 11.0, 18.0, 14.0, 13.0, 22.0, 17.0,
16.0, 12.0, 11.0, 20.0, 16.0, 16.0, 15.0, 18.0, 16.0,
20.0, 22.0, 14.0, 19.0, 21.0, 19.0, 17.0, 15.0, 22.0,
16.0, 22.0, 22.0, 18.0, 21.0],
'fname': 'output\\box(WORDS).png',
'maintitle': 'WORDS',
'val': 'WORDS'}
df=DataFrame()
df.TESTMODE=True
df.read_tbl('data/words~ageXcondition.csv')
D=df.box_plot('WORDS', output_dir='output')
self.assertEqual(D['fname'],R['fname'])
self.assertEqual(D['maintitle'],R['maintitle'])
self.assertEqual(D['val'],R['val'])
for d,r in zip(np.array(D['d']).flat,
np.array(R['d']).flat):
self.assertAlmostEqual(d,r)
def test1(self):
R="""\
t-Test: One Sample for means
SUPPRESSION
=====================================
Sample Mean 19.541
Hypothesized Pop. Mean 17
Variance 228.326
Observations 384
df 383
t Stat 3.295
alpha 0.050
P(T<=t) one-tail 5.384e-04
t Critical one-tail 1.966
P(T<=t) two-tail 0.001
t Critical two-tail 1.649
P(T<=t) two-tail 0.001
Effect size d 0.168
delta 3.295
Observed power one-tail 0.950
Observed power two-tail 0.908 """
df = DataFrame()
df.read_tbl('data/suppression~subjectXgroupXageXcycleXphase.csv')
D=df.ttest('SUPPRESSION', pop_mean=17.)
self.assertEqual(str(D),R)
def test2(self):
df=DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
pt = df.pivot('ERROR', ['MODEL','TIMEOFDAY'],['COURSE'],where=['SUBJECT != 1'])
self.assertEqual(repr(eval(repr(pt))), repr(pt))
def test1(self):
R="""\
Chi-Square: Single Factor
SUMMARY
1 2 3 4
============================================
Observed 7 20 23 9
Expected 14.750 14.750 14.750 14.750
CHI-SQUARE TESTS
Value df P
========================================
Pearson Chi-Square 12.797 3 0.005
Likelihood Ratio 13.288 3 0.004
Observations 59
POST-HOC POWER
Measure
==============================
Effect size w 0.466
Non-centrality lambda 12.797
Critical Chi-Square 7.815
Power 0.865 """
df = DataFrame()
df.read_tbl('data/chi_test.csv')
X=df.chisquare1way('RESULT')
self.assertEqual(str(X),R)
def test11(self):
df = DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
D = str(df.descriptives('ERROR'))
R = """\
Descriptive Statistics
ERROR
==========================
count 48.000
mean 3.896
mode 3.000
var 5.797
stdev 2.408
sem 0.348
rms 4.567
min 0.000
Q1 2.000
median 3.000
Q3 5.000
max 10.000
range 10.000
95ci_lower 3.215
95ci_upper 4.577 """
self.assertEqual(D, R)
def test1(self):
R = Descriptives([('count', 48.0),
('mean', 3.8958333333333335),
('mode', 3.0),
('var', 5.797429078014184),
('stdev', 2.4077850979716158),
('sem', 0.34753384361617046),
('rms', 4.566636252940086),
('min', 0.0),
('Q1', 2.0),
('median', 3.0),
('Q3', 5.0),
('max', 10.0),
('range', 10.0),
('95ci_lower', 3.2146669998456394),
('95ci_upper', 4.5769996668210275)],
cname='ERROR')
df=DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
D=df.descriptives('ERROR')
for k in D.keys():
self.failUnlessAlmostEqual(D[k],R[k])
def export_csv_pivot(request, entidad=1, ano=str(date.today().year)):
consumos = Consumo.objects.filter(entidad__pk=entidad, ano=ano)
from collections import namedtuple
LineaDetalle = namedtuple('LineaDetalle',[u'Año', "Mes", 'Local_o_Vehiculo', "Consumo", "Valor"])
df = DataFrame()
for c in consumos:
if c.content_type.id == 16:
denominacion = Local.objects.get(pk=c.object_id).denominacion
else:
denominacion = Vehiculo.objects.get(pk=c.object_id).denominacion
df.insert(LineaDetalle(c.ano, c.mes, denominacion.encode("utf-8"), c.medida.denominacion.encode("utf-8"), c.valor)._asdict())
pt = df.pivot("Valor", ['Local_o_Vehiculo','Consumo'], ['Mes'])
# get the response object, this can be used as a stream.
response = HttpResponse(mimetype='text/csv')
# force download.
response['Content-Disposition'] = 'attachment;filename=export.csv'
response.write(pt)
return response
def test3(self):
df=DataFrame()
with self.assertRaises(Exception) as cm:
df.box_plot('a', output_dir='output')
self.assertEqual(str(cm.exception),
'Table must have data to print data')
def test1(self):
df=DataFrame()
with self.assertRaises(Exception) as cm:
df.insert([1,2,3,4])
self.assertEqual(str(cm.exception),
'row must be mappable type')
def test12(self):
df=DataFrame()
df['DUM']=range(48) # Shouldn't complain
self.assertEqual(df.keys(),['DUM'])
df['DUM']=['A' for i in range(48)] # Shouldn't complain
self.assertEqual(df.keys(),['DUM'])
self.assertEqual(df._sqltypesdict['DUM'],'text')
def test3(self):
"""unequal"""
df=DataFrame()
df[1]=range(10)
df[2]=range(10)
df[3]=range(10)
df[4]=range(9)
self.assertFalse(df._are_col_lengths_equal())
def test0(self):
## Within test
df=DataFrame()
fname='error~subjectXtimeofdayXcourseXmodel.csv'
df.read_tbl(fname)
aov=Anova()
aov.run(df,'ERROR',wfactors=['TIMEOFDAY','COURSE','MODEL'])#,transform='windsor05')
aov.output2html(fname[:-4]+'RESULTS.htm')
def test01(self):
df=DataFrame()
df.TESTMODE = True
df.read_tbl('data/iqbrainsize.txt', delimiter='\t')
D = df.scatter_plot('TOTVOL','FIQ',
output_dir='output')
self.assertEqual(None, D['trend'])
def test1(self):
df=DataFrame()
df.TESTMODE=True
df.read_tbl('data/words~ageXcondition.csv')
D=df.histogram_plot('WORDS', cumulative=True, output_dir='output')
self.assertEqual(D['fname'],'output\\hist(WORDS,cumulative=True).png')
def test2(self):
"""equal non-zero"""
df=DataFrame()
df[1]=range(10)
df[2]=range(10)
df[3]=range(10)
df[4]=range(10)
self.assertTrue(df._are_col_lengths_equal())
def test3(self):
## Mixed Between/Within test
df=DataFrame()
fname='suppression~subjectXgroupXcycleXphase.csv'
df.read_tbl(fname)
df['SUPPRESSION']=[.01*x for x in df['SUPPRESSION']]
aov=Anova()
aov.run(df,'SUPPRESSION',wfactors=['CYCLE','PHASE'],bfactors=['GROUP'])#,transform='win
def test05(self):
R = """Marginals([('factorials', OrderedDict([('AGE', [u'old', u'old', u'old', u'old', u'old']), ('CONDITION', [u'adjective', u'counting', u'imagery', u'intention', u'rhyming'])])), ('dmu', [11.0, 7.0, 13.4, 12.0, 6.9000000000000004]), ('dN', [10, 10, 10, 10, 10]), ('dsem', [0.78881063774661542, 0.57735026918962573, 1.4236104336041748, 1.1832159566199232, 0.67412494720522276]), ('dlower', [9.4539311500166345, 5.868393472388334, 10.609723550135818, 9.6808967250249509, 5.578715103477764]), ('dupper', [12.546068849983365, 8.131606527611666, 16.190276449864182, 14.319103274975049, 8.2212848965222367])], val='WORDS', factors=['AGE', 'CONDITION'], where='AGE == "old"')"""
df=DataFrame()
df.read_tbl('data/words~ageXcondition.csv')
D = df.marginals('WORDS',
factors=['AGE','CONDITION'],
where='AGE == "old"')
def test3(self):
df=DataFrame()
with self.assertRaises(Exception) as cm:
df.validate({'GROUP' : lambda x: x in ['AA', 'AB', 'LAB']})
self.assertEqual(str(cm.exception),
'table must have data to validate data')
def test3(self):
df=DataFrame()
with self.assertRaises(Exception) as cm:
df.sort()
self.assertEqual(str(cm.exception),
'Table must have data to sort data')
def test3(self):
R = {'aggregate': 'ci',
'clevels': ['I', 'II'],
'fname': 'output\\whereGROUPnotLAB.png',
'maintitle': 'SUPPRESSION by CYCLE * AGE * PHASE * GROUP',
'numcols': 2,
'numrows': 2,
'rlevels': ['AA', 'AB'],
'subplot_titles': ['GROUP = AA, PHASE = AA',
'GROUP = AA, PHASE = AA',
'GROUP = AB, PHASE = AB',
'GROUP = AB, PHASE = AB'],
'xmaxs': [4.1500000000000004,
4.1500000000000004,
4.1500000000000004,
4.1500000000000004],
'xmins': [0.84999999999999998,
0.84999999999999998,
0.84999999999999998,
0.84999999999999998],
'y': [[[ 17.75 , 22.375, 23.125, 20.25 ],
[ 8.675, 10.225, 10.5 , 9.925]],
[[ 20.875, 28.125, 20.75 , 24.25 ],
[ 8.3 , 10.25 , 9.525, 11.1 ]],
[[ 12.625, 23.5 , 20. , 15.625],
[ 5.525, 8.825, 9.125, 7.75 ]],
[[ 22.75 , 41.125, 46.125, 51.75 ],
[ 8.675, 13.1 , 14.475, 12.85 ]]],
'ymax': 64.8719707118471,
'ymin': 0.0}
# separate y plots and separate x plots
df=DataFrame()
df.TESTMODE = True
df.read_tbl('data\suppression~subjectXgroupXageXcycleXphase.csv')
D = df.interaction_plot('SUPPRESSION','CYCLE',
seplines='AGE',
sepxplots='PHASE',
sepyplots='GROUP',yerr='ci',
where=[('GROUP','not in',['LAB'])],
fname='whereGROUPnotLAB.png',
output_dir='output')
self.assertEqual(D['aggregate'], R['aggregate'])
self.assertEqual(D['clevels'], R['clevels'])
self.assertEqual(D['rlevels'], R['rlevels'])
self.assertEqual(D['numcols'], R['numcols'])
self.assertEqual(D['numrows'], R['numrows'])
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['subplot_titles'], R['subplot_titles'])
self.assertAlmostEqual(D['ymin'], R['ymin'])
self.assertAlmostEqual(D['ymax'], R['ymax'])
for d,r in zip(np.array(D['y']).flat,np.array(R['y']).flat):
self.assertAlmostEqual(d,r)
def test1(self):
R = """\
CYCLE PHASE GROUP=AA, GROUP=AA, GROUP=AB, GROUP=AB, GROUP=LAB, GROUP=LAB,
AGE=old AGE=young AGE=old AGE=young AGE=old AGE=young
=======================================================================================
1 I 17.750 8.675 12.625 5.525 21.625 7.825
1 II 20.875 8.300 22.750 8.675 36.250 13.750
2 I 22.375 10.225 23.500 8.825 21.375 9.900
2 II 28.125 10.250 41.125 13.100 46.875 14.375
3 I 23.125 10.500 20.000 9.125 23.750 9.500
3 II 20.750 9.525 46.125 14.475 50.375 15.575
4 I 20.250 9.925 15.625 7.750 26.375 9.650
4 II 24.250 11.100 51.750 12.850 46.500 14.425 """
df = DataFrame()
df.read_tbl('data/suppression~subjectXgroupXageXcycleXphase.csv')
pt = df.pivot('SUPPRESSION',
rows=['CYCLE', 'PHASE'],
cols=['GROUP', 'AGE'])
df2 = pt.to_dataframe()
self.assertEqual(str(df2), R)
def test01(self):
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data/iqbrainsize.txt', delimiter='\t')
D = df.scatter_plot('TOTVOL', 'FIQ', output_dir='output')
self.assertEqual(None, D['trend'])
def test1(self):
R = {
'd': [
np.array([
9, 8, 6, 8, 10, 4, 6, 5, 7, 7, 7, 9, 6, 6, 6, 11, 6, 3, 8,
7, 11, 13, 8, 6, 14, 11, 13, 13, 10, 11, 12, 11, 16, 11, 9,
23, 12, 10, 19, 11, 10, 19, 14, 5, 10, 11, 14, 15, 11, 11
]),
np.array([
8, 6, 4, 6, 7, 6, 5, 7, 9, 7, 10, 7, 8, 10, 4, 7, 10, 6, 7,
7, 14, 11, 18, 14, 13, 22, 17, 16, 12, 11, 20, 16, 16, 15,
18, 16, 20, 22, 14, 19, 21, 19, 17, 15, 22, 16, 22, 22, 18,
21
])
],
'fname':
'output\\box(WORDS~AGE).png',
'maintitle':
'WORDS by AGE',
'xlabels': ['AGE = old', 'AGE = young']
}
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data/words~ageXcondition.csv')
D = df.box_plot('WORDS', ['AGE'], output_dir='output')
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['xlabels'], R['xlabels'])
for d, r in zip(np.array(D['d']).flat, np.array(R['d']).flat):
self.assertAlmostEqual(d, r)
def test0(self):
R = {
'd': [
9.0, 8.0, 6.0, 8.0, 10.0, 4.0, 6.0, 5.0, 7.0, 7.0, 7.0, 9.0,
6.0, 6.0, 6.0, 11.0, 6.0, 3.0, 8.0, 7.0, 11.0, 13.0, 8.0, 6.0,
14.0, 11.0, 13.0, 13.0, 10.0, 11.0, 12.0, 11.0, 16.0, 11.0,
9.0, 23.0, 12.0, 10.0, 19.0, 11.0, 10.0, 19.0, 14.0, 5.0, 10.0,
11.0, 14.0, 15.0, 11.0, 11.0, 8.0, 6.0, 4.0, 6.0, 7.0, 6.0,
5.0, 7.0, 9.0, 7.0, 10.0, 7.0, 8.0, 10.0, 4.0, 7.0, 10.0, 6.0,
7.0, 7.0, 14.0, 11.0, 18.0, 14.0, 13.0, 22.0, 17.0, 16.0, 12.0,
11.0, 20.0, 16.0, 16.0, 15.0, 18.0, 16.0, 20.0, 22.0, 14.0,
19.0, 21.0, 19.0, 17.0, 15.0, 22.0, 16.0, 22.0, 22.0, 18.0,
21.0
],
'fname':
'output\\box(WORDS).png',
'maintitle':
'WORDS',
'val':
'WORDS'
}
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data/words~ageXcondition.csv')
D = df.box_plot('WORDS', output_dir='output')
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['val'], R['val'])
for d, r in zip(np.array(D['d']).flat, np.array(R['d']).flat):
self.assertAlmostEqual(d, r)
def test2(self):
R = {
'A': [
-8.0, -7.0, -3.0, -2.0, -1.0, 1.0, 2.0, 3.0, 4.0, 9.0, -10.0,
-9.0, -6.0, -5.0, -4.0, 0.0, 5.0, 6.0, 7.0, 8.0
],
'B': [
1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 2.0, 2.0,
2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0, 2.0
]
}
a = [
4, 8, 1, 5, -7, -5, 9, 7, -8, -10, -1, -4, 3, 0., -2, 6, 2, -9, -3,
-6
]
b = [1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2, 1, 2]
df = DataFrame()
for A, B in zip(a, b):
df.insert({'A': A, 'B': B})
df.sort(['B', 'A'])
for d, r in zip(df['A'], R['A']):
self.assertAlmostEqual(d, r)
for d, r in zip(df['B'], R['B']):
self.assertAlmostEqual(d, r)
class Test_writeTable(unittest.TestCase):
def setUp(self):
self.df = DataFrame()
self.df.read_tbl('data/suppression~subjectXgroupXageXcycleXphase.csv')
def test0(self):
d = 'data/suppression~subjectXgroupXageXcycleXphase.csv'
r = 'subjectXsexXageXgroupXcycleXphaseXsuppressionXranddata.csv'
self.df.write()
self.assertTrue(fcmp(d, r))
# clean up
os.remove(
'./subjectXsexXageXgroupXcycleXphaseXsuppressionXranddata.csv')
def test1(self):
# with exclusion
d = 'data/suppression~subjectXgroupXageXcycleXphase.csv'
r = 'subjectXsexXageXgroupXcycleXphaseXsuppressionXranddata.csv'
self.df.write(where=[('AGE', 'not in', ['young'])])
self.assertTrue(fcmp(d, r))
# clean up
os.remove(
'./subjectXsexXageXgroupXcycleXphaseXsuppressionXranddata.csv')
def test06(self):
R = """\
y 1 y 2 y 3
===============
1 5 9
2 6 --
3 7 11
4 8 12 """
# labels have spaces
with open('test.csv', 'wb') as f:
f.write("""
y 1,y 2,y 3
1,5,9
2,6,
3,7,11
4,8,12""")
self.df = DataFrame()
self.df.read_tbl('test.csv', skip=1, labels=True)
self.assertAlmostEqual(str(self.df), R)
def test05(self):
R = """\
x y z
==========
1 5 9
2 6 --
3 7 11
4 8 12 """
# cell has empty string, comma after 6
with open('test.csv', 'wb') as f:
f.write("""
x,y,z
1,5,9
2,6,
3,7,11
4,8,12""")
self.df = DataFrame()
self.df.read_tbl('test.csv', skip=1, labels=True)
self.assertAlmostEqual(str(self.df), R)
def test4(self):
df = DataFrame()
df.insert([('GROUP', 'AA'), ('VAL', 1)])
with self.assertRaises(Exception) as cm:
df.validate(lambda x: x in ['AA', 'AB', 'LAB'])
self.assertEqual(str(cm.exception), 'criteria must be mappable type')
def test1(self):
R = DataFrame([('SUBJECT', [1, 2]), ('TIMEOFDAY', ['T1', 'T1']),
('COURSE', ['C1', 'C2']), ('MODEL', ['M1', 'M1']),
('ERROR', [10, 10])])
df = DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
df2 = df.where(['ERROR = 10'])
self.assertEqual(repr(df2), repr(R))
def test1(self):
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data/words~ageXcondition.csv')
D = df.histogram_plot('WORDS', cumulative=True, output_dir='output')
self.assertEqual(D['fname'], 'output\\hist(WORDS,cumulative=True).png')
def test3(self):
R = """c b{L@^hsa aj} a(1%32@)
===========================
1 a 34
2 b 34
3 c 42
4 d 34
5 e 45
6 f 34 """
df=DataFrame()
df.PRINTQUERIES = True
df.insert({'a(1%32@)':34,'b{L@^hsa aj}':'a','c':1})
df.insert({'a(1%32@)':34,'b{L@^hsa aj}':'b','c':2})
df.insert({'a(1%32@)':42,'b{L@^hsa aj}':'c','c':3})
df.insert({'a(1%32@)':34,'b{L@^hsa aj}':'d','c':4})
df.insert({'a(1%32@)':45,'b{L@^hsa aj}':'e','c':5})
df.insert({'a(1%32@)':34,'b{L@^hsa aj}':'f','c':6})
self.assertEqual(R, str(df))
def test2(self):
R="""Anova: Single Factor on SUPPRESSION
SUMMARY
Groups Count Sum Average Variance
==============================================
AA 128 2048 16 148.792
AB 128 2510.600 19.614 250.326
LAB 128 2945.000 23.008 264.699
ANOVA
Source of SS df MS F P-value
Variation
===========================================================
Treatments 3144.039 2 1572.020 7.104 9.348e-04
Error 84304.687 381 221.272
===========================================================
Total 87448.726 383 """
df = DataFrame()
df.read_tbl('suppression~subjectXgroupXageXcycleXphase.csv')
aov=df.anova1way('SUPPRESSION','GROUP')
self.assertEqual(str(aov),R)
def test3(self):
df = DataFrame()
df.insert({'A': 1, 'B': 2})
with self.assertRaises(Exception) as cm:
df.insert({'A': 1, 'B': 2, 'C': 3})
self.assertEqual(str(cm.exception),
'row must have the same keys as the table')
def test2(self):
R = {
'aggregate':
'ci',
'clevels': [1],
'fname':
'output\\interaction_plot(SUPPRESSION~CYCLE_X_AGE_X_PHASE,yerr=95% ci).png',
'maintitle':
'SUPPRESSION by CYCLE * AGE * PHASE',
'numcols':
1,
'numrows':
2,
'rlevels': ['I', 'II'],
'subplot_titles': ['I', 'II'],
'xmaxs': [4.1749999999999998, 4.1749999999999998],
'xmins': [0.32499999999999996, 0.32499999999999996],
'y': [[[17.33333333, 22.41666667, 22.29166667, 20.75],
[7.34166667, 9.65, 9.70833333, 9.10833333]],
[[26.625, 38.70833333, 39.08333333, 40.83333333],
[10.24166667, 12.575, 13.19166667, 12.79166667]]],
'yerr': [[1.81325589, 1.44901936, 1.60883063, 1.57118871],
[2.49411239, 1.34873573, 1.95209851, 1.35412572]],
'ymax':
64.8719707118471,
'ymin':
0.0
}
# generate yerr
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data\suppression~subjectXgroupXageXcycleXphase.csv')
D = df.interaction_plot('SUPPRESSION',
'CYCLE',
seplines='AGE',
sepyplots='PHASE',
yerr='ci',
output_dir='output')
self.assertEqual(D['aggregate'], R['aggregate'])
self.assertEqual(D['clevels'], R['clevels'])
self.assertEqual(D['rlevels'], R['rlevels'])
self.assertEqual(D['numcols'], R['numcols'])
self.assertEqual(D['numrows'], R['numrows'])
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['subplot_titles'], R['subplot_titles'])
self.assertAlmostEqual(D['ymin'], R['ymin'])
self.assertAlmostEqual(D['ymax'], R['ymax'])
for d, r in zip(np.array(D['y']).flat, np.array(R['y']).flat):
self.assertAlmostEqual(d, r)
for d, r in zip(np.array(D['yerr']).flat, np.array(R['yerr']).flat):
self.assertAlmostEqual(d, r)
def test0(self):
R = """Chi-Square: two Factor
SUMMARY
Guilty NotGuilty Total
=====================================
High 105 76 181
(130.441) (50.559)
Low 153 24 177
(127.559) (49.441)
=====================================
Total 258 100 358
SYMMETRIC MEASURES
Value Approx.
Sig.
===========================================
Cramer's V 0.317 8.686e-10
Contingency Coefficient 0.302 5.510e-09
N of Valid Cases 358
CHI-SQUARE TESTS
Value df P
===============================================
Pearson Chi-Square 35.930 1 2.053e-09
Continuity Correction 34.532 1 4.201e-09
Likelihood Ratio 37.351 1 0
N of Valid Cases 358 """
df = DataFrame()
df['FAULTS'] = list(Counter(Low=177, High=181).elements())
df['FAULTS'].reverse()
df['VERDICT'] = list(Counter(Guilty=153, NotGuilty=24).elements())
df['VERDICT'].extend(list(
Counter(Guilty=105, NotGuilty=76).elements()))
x2 = df.chisquare2way('FAULTS', 'VERDICT')
self.assertEqual(str(x2), R)
def test02(self):
df = DataFrame()
df.read_tbl('data/words~ageXcondition.csv')
D = str(df.descriptives('WORDS'))
R = """\
Descriptive Statistics
WORDS
==========================
count 100.000
mean 11.610
mode 11.000
var 26.947
stdev 5.191
sem 0.519
rms 12.707
min 3.000
Q1 7.000
median 11.000
Q3 15.500
max 23.000
range 20.000
95ci_lower 10.593
95ci_upper 12.627 """
self.assertEqual(D, R)
def test5(self):
R = [
"""\
avg(ERROR)
COURSE TIMEOFDAY=T1 TIMEOFDAY=T2
====================================
C1 7.167 3.222 """, """\
avg(ERROR)
COURSE TIMEOFDAY=T1 TIMEOFDAY=T2
====================================
C2 6.500 2.889 """, """\
avg(ERROR)
COURSE TIMEOFDAY=T1 TIMEOFDAY=T2
====================================
C3 4 1.556 """
]
df = DataFrame()
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
pt = df.pivot('ERROR', ['TIMEOFDAY'], ['COURSE'])
for r, L in zip(R, pt.transpose()):
self.assertAlmostEqual(r, str(L))
def test1(self):
R = {
'aggregate':
None,
'clevels': ['M1', 'M2', 'M3'],
'fname':
'output\\interaction_plot(ERROR~TIMEOFDAY_X_COURSE_X_MODEL,yerr=1.0).png',
'maintitle':
'ERROR by TIMEOFDAY * COURSE * MODEL',
'numcols':
3,
'numrows':
1,
'rlevels': [1],
'subplot_titles': ['M1', 'M2', 'M3'],
'xmaxs': [1.5, 1.5, 1.5],
'xmins': [-0.5, -0.5, -0.5],
'y': [[[9., 4.33333333], [8.66666667, 3.66666667],
[4.66666667, 1.66666667]],
[[7.5, 2.66666667], [6., 2.66666667], [5., 1.66666667]],
[[5., 2.66666667], [3.5, 2.33333333],
[2.33333333, 1.33333333]]],
'yerr': [[1.0, 1.0], [1.0, 1.0], [1.0, 1.0]],
'ymax':
11.119188627248182,
'ymin':
0.0
}
# specify yerr
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data/error~subjectXtimeofdayXcourseXmodel_MISSING.csv')
D = df.interaction_plot('ERROR',
'TIMEOFDAY',
seplines='COURSE',
sepxplots='MODEL',
yerr=1.,
output_dir='output')
self.assertEqual(D['aggregate'], R['aggregate'])
self.assertEqual(D['clevels'], R['clevels'])
self.assertEqual(D['rlevels'], R['rlevels'])
self.assertEqual(D['numcols'], R['numcols'])
self.assertEqual(D['numrows'], R['numrows'])
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['subplot_titles'], R['subplot_titles'])
self.assertAlmostEqual(D['ymin'], R['ymin'])
self.assertAlmostEqual(D['ymax'], R['ymax'])
for d, r in zip(np.array(D['y']).flat, np.array(R['y']).flat):
self.assertAlmostEqual(d, r)
for d, r in zip(np.array(D['yerr']).flat, np.array(R['yerr']).flat):
self.assertAlmostEqual(d, r)
def test2(self):
R = """\
Chi-Square: Single Factor
SUMMARY
1 2 3 4 5
=====================================================
Observed 7 20 23 9 0
Expected 11.800 11.800 11.800 11.800 11.800
CHI-SQUARE TESTS
Value df P
============================================
Pearson Chi-Square 30.746 4 3.450e-06
Likelihood Ratio -- -- --
Observations 59
POST-HOC POWER
Measure
==============================
Effect size w 0.722
Non-centrality lambda 30.746
Critical Chi-Square 9.488
Power 0.998 """
df = DataFrame()
df.read_tbl('data/chi_test.csv')
X = df.chisquare1way('RESULT', {
1: 11.8,
2: 11.8,
3: 11.8,
4: 11.8,
5: 11.8
})
self.assertEqual(str(X), R)
def test4(self):
R = """t-Test: Paired Two Sample for means
PRE POST
=============================================
Mean 87.250 87.083
Variance 1207.659 1166.629
Observations 12 12
Pearson Correlation 0.995
df 11
t Stat 0.163
alpha 0.050
P(T<=t) one-tail 0.437
t Critical one-tail 2.201
P(T<=t) two-tail 0.873
t Critical two-tail 1.796
P(T<=t) two-tail 0.873
Effect size dz 0.047
delta 0.163
Observed power one-tail 0.068
Observed power two-tail 0.035 """
df = DataFrame()
df.read_tbl('data/example2_prepost.csv')
D = df.ttest('PRE', 'POST', paired=True)
self.assertEqual(str(D), R)
def test3(self):
tupa = ('a1','a2','a3')
tupb = ('a1','b2','b3')
df=DataFrame()
df.insert([(('a1','a2','a3'),34), (('a1','b2','b3'),1)])
df.insert([(('a1','a2','a3'),34), (('a1','b2','b3'),2)])
df.insert([(('a1','a2','a3'),42), (('a1','b2','b3'),3)])
namea,nameb = df.keys()
self.assertEqual(namea, tupa)
self.assertEqual(nameb, tupb)
def test01(self):
"""confidence interval error bars specified"""
R = {
'aggregate':
'ci',
'clevels': [1],
'fname':
'output\\interaction_plot(WORDS~AGE_X_CONDITION,yerr=95% ci).png',
'maintitle':
'WORDS by AGE * CONDITION',
'numcols':
1,
'numrows':
1,
'rlevels': [1],
'subplot_titles': [''],
'xmaxs': [1.5],
'xmins': [-0.5],
'y': [[[11.0, 14.8], [7.0, 6.5], [13.4, 17.6], [12.0, 19.3],
[6.9, 7.6]]],
'yerr': [[]],
'ymin':
0.0,
'ymax':
27.183257964740832
}
# a simple plot
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data/words~ageXcondition.csv')
D = df.interaction_plot('WORDS',
'AGE',
seplines='CONDITION',
output_dir='output',
yerr='ci')
self.assertEqual(D['aggregate'], R['aggregate'])
self.assertEqual(D['clevels'], R['clevels'])
self.assertEqual(D['rlevels'], R['rlevels'])
self.assertEqual(D['numcols'], R['numcols'])
self.assertEqual(D['numrows'], R['numrows'])
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['subplot_titles'], R['subplot_titles'])
self.assertAlmostEqual(D['ymin'], R['ymin'])
self.assertAlmostEqual(D['ymax'], R['ymax'])
for d, r in zip(np.array(D['y']).flat, np.array(R['y']).flat):
self.assertAlmostEqual(d, r)
for d, r in zip(np.array(D['yerr']).flat, np.array(R['yerr']).flat):
self.assertAlmostEqual(d, r)
def test1(self):
"""chi-square 2-way"""
R = """\
Chi-Square: two Factor
SUMMARY
Litter Removed Trash Can Total
====================================================
Countrol 385 477 41 903
(343.976) (497.363) (61.661)
Message 290 499 80 869
(331.024) (478.637) (59.339)
====================================================
Total 675 976 121 1772
SYMMETRIC MEASURES
Value Approx.
Sig.
===========================================
Cramer's V 0.121 3.510e-07
Contingency Coefficient 0.120 4.263e-07
N of Valid Cases 1772
CHI-SQUARE TESTS
Value df P
============================================
Pearson Chi-Square 25.794 2 2.506e-06
Likelihood Ratio 26.056 2 2.198e-06
N of Valid Cases 1772
CHI-SQUARE POST-HOC POWER
Measure
==============================
Effect size w 0.121
Non-centrality lambda 25.794
Critical Chi-Square 5.991
Power 0.997 """
df = DataFrame()
rfactors = ['Countrol'] * 903 + ['Message'] * 869
cfactors = ['Trash Can'] * 41 + ['Litter'] * 385 + ['Removed'] * 477
cfactors += ['Trash Can'] * 80 + ['Litter'] * 290 + ['Removed'] * 499
x2 = ChiSquare2way()
x2.run(rfactors, cfactors)
self.assertEqual(str(x2), R)
def test4(self):
R = {
'aggregate':
None,
'clevels':
['adjective', 'counting', 'imagery', 'intention', 'rhyming'],
'fname':
'output\\interaction_plot(WORDS~AGE_X_CONDITION).png',
'maintitle':
'WORDS by AGE * CONDITION',
'numcols':
5,
'numrows':
1,
'rlevels': [1],
'subplot_titles':
['adjective', 'counting', 'imagery', 'intention', 'rhyming'],
'xmaxs': [1.5, 1.5, 1.5, 1.5, 1.5],
'xmins': [-0.5, -0.5, -0.5, -0.5, -0.5],
'y': [[11., 14.8], [7., 6.5], [13.4, 17.6], [12., 19.3],
[6.9, 7.6]],
'yerr': [[], [], [], [], []],
'ymax':
27.183257964740832,
'ymin':
0.0
}
# a simple plot
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data/words~ageXcondition.csv')
D = df.interaction_plot('WORDS',
'AGE',
sepxplots='CONDITION',
output_dir='output')
self.assertEqual(D['aggregate'], R['aggregate'])
self.assertEqual(D['clevels'], R['clevels'])
self.assertEqual(D['rlevels'], R['rlevels'])
self.assertEqual(D['numcols'], R['numcols'])
self.assertEqual(D['numrows'], R['numrows'])
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['subplot_titles'], R['subplot_titles'])
self.assertAlmostEqual(D['ymin'], R['ymin'])
self.assertAlmostEqual(D['ymax'], R['ymax'])
for d, r in zip(np.array(D['y']).flat, np.array(R['y']).flat):
self.assertAlmostEqual(d, r)
for d, r in zip(np.array(D['yerr']).flat, np.array(R['yerr']).flat):
self.assertAlmostEqual(d, r)
def test6(self):
R = {
'aggregate':
'ci',
'clevels': [1],
'fname':
'output\\interaction_plot(SUPPRESSION~CYCLE_X_PHASE,yerr=95% ci).png',
'maintitle':
'SUPPRESSION by CYCLE * PHASE',
'numcols':
1,
'numrows':
2,
'rlevels': ['I', 'II'],
'subplot_titles': ['I', 'II'],
'xmaxs': [4.1749999999999998, 4.1749999999999998],
'xmins': [0.82499999999999996, 0.82499999999999996],
'y': [[12.3375, 16.03333333, 16., 14.92916667],
[18.43333333, 25.64166667, 26.1375, 26.8125]],
'yerr': [[3.18994762, 3.20528834, 3.26882751, 3.53477953],
[3.98429064, 4.5950803, 4.9514978, 4.97429769]],
'ymax':
64.8719707118471,
'ymin':
0.0
}
# generate yerr
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data\suppression~subjectXgroupXageXcycleXphase.csv')
D = df.interaction_plot('SUPPRESSION',
'CYCLE',
sepyplots='PHASE',
yerr='ci',
output_dir='output')
self.assertEqual(D['aggregate'], R['aggregate'])
self.assertEqual(D['clevels'], R['clevels'])
self.assertEqual(D['rlevels'], R['rlevels'])
self.assertEqual(D['numcols'], R['numcols'])
self.assertEqual(D['numrows'], R['numrows'])
self.assertEqual(D['fname'], R['fname'])
self.assertEqual(D['maintitle'], R['maintitle'])
self.assertEqual(D['subplot_titles'], R['subplot_titles'])
self.assertAlmostEqual(D['ymin'], R['ymin'])
self.assertAlmostEqual(D['ymax'], R['ymax'])
for d, r in zip(np.array(D['y']).flat, np.array(R['y']).flat):
self.assertAlmostEqual(d, r)
for d, r in zip(np.array(D['yerr']).flat, np.array(R['yerr']).flat):
self.assertAlmostEqual(d, r)
def test31(self):
# separate y plots and separate x plots
df = DataFrame()
df.TESTMODE = True
df.read_tbl('data\suppression~subjectXgroupXageXcycleXphase.csv')
D = df.interaction_plot('SUPPRESSION',
'CYCLE',
seplines='AGE',
sepxplots='GROUP',
sepyplots='PHASE',
yerr='sem',
output_dir='output')
def test22(self):
"""test with string keys and where condition"""
df = DataFrame()
df['1'] = list(range(100))
df['2'] = ['bob' for i in range(100)]
df['3'] = [i * 1.234232 for i in range(100)]
df['4'] = ['bob' for i in range(50)] + list(range(50))
shuffle(df['1'])
shuffle(df['2'])
shuffle(df['3'])
df._build_sqlite3_tbl(list(df.keys())[:2], ['4 not in ("bob")'])
df._execute('select * from TBL')
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df['1'][i + 50])
self.assertEqual(b, df['2'][i + 50])
def test3(self):
"""test with string keys and tuple where condition"""
df = DataFrame()
df[1] = list(range(100))
df[2] = ['bob' for i in range(100)]
df[3] = [i * 1.234232 for i in range(100)]
df[4] = ['bob' for i in range(50)] + list(range(50))
shuffle(df[1])
shuffle(df[2])
shuffle(df[3])
df._build_sqlite3_tbl(list(df.keys())[:2], [(4, '!=', 'bob')])
df._execute('select * from TBL')
for i, (a, b) in enumerate(df.cur):
self.assertEqual(a, df[1][i + 50])
self.assertEqual(b, df[2][i + 50])
def long2wide(in_fname,
id,
dvs,
between=[],
within=[],
covariates=[],
out_fname=None,
nested=True):
# load in_fname into a PyvtTbl object
print(('reading "%s"...' % in_fname))
cls = DataFrame()
cls.read_tbl(in_fname)
# loop through DVs and append within columns
d = [sorted(set(cls[id]))]
header = [id] + covariates + between
for col in covariates + between:
z = cls.pivot(col, cols=[id], aggregate='arbitrary')
d.extend(list(z))
# start controls whether nested factors are examined
if nested: start = 1
else: start = len(within)
for i, dv in enumerate(dvs):
print(('\ncollaborating %s' % dv))
for j in _xrange(start, len(within) + 1):
for factors in _xunique_combinations(within, j):
print((' pivoting', factors, '...'))
z = cls.pivot(dv, rows=factors, cols=[id], aggregate='avg')
d.extend(list(z))
# process headers
for names in z.rnames:
h = '_'.join(('%s.%s' % (f, str(c)) for (f, c) in names))
header.append('%s__%s' % (dv, h))
# Now we can write the data
if out_fname == None:
out_fname = 'wide_data.csv'
with open(out_fname, 'wb') as f:
wtr = csv.writer(f)
wtr.writerow([n.upper() for n in header])
wtr.writerows(list(zip(*d))) # transpose and write
