pysmile/example_code2.py at master · kadeng/pysmile · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142

#Testing the functionality of the pysmile library. This library is a direct wrapper around the SMILE and SMILEARN Libraries which belong to the GeNiE Bayesian Network Software. The Library is documented here: <a href="http://genie.sis.pitt.edu/wiki/Main_Page">GeNiE & Smile Documentation Wiki</a>

from pysmile import *
# Basic int array functionality
ia2 = DSL_intArray(20)
items = intArray.frompointer(ia2.Items()) # Warning: items is an *unchecked* array of ints now.
# We can even do pointer stuff
for i in range(0,20):
    items[i] = i+6

h = intPointer.frompointer(ia2.Subscript(10))

h.assign(21)
# Basic double array functionality
da2 = DSL_doubleArray(20)
ditems = doubleArray.frompointer(da2.Items()) # Warning: items is an *unchecked* array of doubles now.

for i in range(0,20):
    ditems[i] = i+6.4

ditems[10]
# Basic string array functionality
sa2 = DSL_stringArray(20,10)

for i in range(0,20):
    sa2.SetString(i, "Alles klar %d" % (i))

print sa2.Subscript(10)
# Basic id array functionality
ia2 = DSL_idArray(20,10)

for i in range(0,20):
    ia2.SetString(i, "AllesKlar%d" % (i))

print ia2.Subscript(10)
ia2.SetString(4, "Alles Klar %d" % (4))

print ia2.Subscript(4)
ar = toDSL_intArray([1,2,3])
ari = intArray.frompointer(ar.Items())

ar = toDSL_doubleArray([1.1,2.2,3.3])
ari = doubleArray.frompointer(ar.Items())
dm = DSL_Dmatrix(toDSL_intArray([2,2,3]))
import pysmile
theNet = DSL_network()
success = theNet.AddNode(DSL_CPT,"Success")
someNames = DSL_stringArray()
someNames.Add("Success")
someNames.Add("Failure")
theNet.GetNode(success).Definition().SetNumberOfOutcomes(someNames)
forecast = theNet.AddNode(DSL_CPT,"Forecast")
someNames.Flush()
someNames.Add("Good")
someNames.Add("Moderate")
someNames.Add("Poor")
theNet.GetNode(forecast).Definition().SetNumberOfOutcomes(someNames)
theNet.AddArc(success,forecast)
theProbs = DSL_doubleArray()
theProbs.SetSize(2)
theProbs.setUnchecked(1, 0.3)
theProbsItem = doubleArray.frompointer(theProbs.Items())
theProbsItem[0] = 0.7
theNet.GetNode(success).Definition().SetDefinition(theProbs);
theNet.WriteFile("tutorial.xdsl")
theCoordinates = DSL_sysCoordinates(theNet.GetNode(forecast).Definition());
theCoordinates.GoTo(0)
ds = DSL_dataset()
ds.ReadFile('/data/fast/ltm_wifi_test/NISLTM/train_set.csv')
ds.GetNumberOfRecords()
ds.GetNumberOfVariables()
learner = DSL_tan()
learnedTANNetwork = DSL_network()
learner.maxSearchTime=1000
learner.classvar = "wifi_usage"
learner.Learn(ds, learnedTANNetwork)
learnedTANNetwork.WriteFile("tan_learned1.xdsl")
from pysmile import *

ds = DSL_dataset()
ds.ReadFile('/data/fast/ltm_wifi_test/NISLTM/train_set.csv')

# Learn a Tree Augmented Naived Bayes Model
learner = DSL_tan()
learnedTANNetwork = DSL_network()
learner.maxSearchTime=int(1000)
learner.classvar = "wifi_usage"
learner.Learn(ds, learnedTANNetwork)
learnedTANNetwork.WriteFile("tan_learned.xdsl")

learnedTANNetwork.GetNode(learnedTANNetwork.FindNode("domain_content")
vds = DSL_dataset()
vds.ReadFile('validation_set.csv')
vds.RemoveVar(vds.FindVariable("wifi_usage"))
vds.RemoveVar(vds.FindVariable("domain_content"))
vds.RemoveVar(vds.FindVariable("domain_context"))
vds.RemoveVar(vds.FindVariable("domain_activity"))
vds.RemoveVar(vds.FindVariable("device_wifi_usage_class"))

vmatches = DSL_datasetMatch_vector()
vds.MatchNetworkSimple(learnedTANNetwork, vmatches)
case = DSL_simpleCase(learnedTANNetwork)
case.SetEvidenceByHandle(learnedTANNetwork.FindNode("age"), 1)
case.CaseToNetwork()
learnedTANNetwork.UpdateBeliefs()
wifi_node = learnedTANNetwork.GetNode(learnedTANNetwork.FindNode("age"))
val = wifi_node.Value()
values = val.GetMatrix()
va = doubleArray.frompointer(wifi_node.Value().GetMatrix().GetItems().Items())
va[1]
learnedTANNetwork.ClearAllEvidence()
learnedTANNetwork.UpdateBeliefs()
age_node = learnedTANNetwork.GetNode(learnedTANNetwork.FindNode("age"))
val = age_node.Value()
val.SetEvidence(1)
values = val.GetMatrix()
va[0]
ds2 = DSL_dataset()
ds2.ReadFile('/data/fast/ltm_wifi_test/NISLTM/train_set.csv')
for name in ds2.GetStateNames(ds2.FindVariable("domain_content")):
    print name

for name in vds.GetStateNames(vds.FindVariable("domain_content")):
    print name

x = DSNetworkMatch(learnedTANNetwork)
print x.attachDataset(vds)
wifi_usage_node = learnedTANNetwork.FindNode("wifi_usage")
learnedTANNetwork.ClearAllEvidence()
learnedTANNetwork.UpdateBeliefs()
va = doubleArray.frompointer(learnedTANNetwork.GetNode(wifi_usage_node).Value().GetMatrix().GetItems().Items())
print va[0]

for i in range(10):
    x.enterEvidenceFromDatasetRow(21*i)
    learnedTANNetwork.UpdateBeliefs()
    print va[0]

outc = ndef.GetOutcomesNames()
for i in range(outc.GetSize()):
    print outc.Subscript(i)