def setUp(self):
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
dataDir = PathDefaults.getDataDir() + "infoDiffusion/"
#dataDir = "/home/charanpal/Private/Postdoc/Code/APGL/data/"
matFileName = dataDir + "EgoAlterTransmissions.mat"
self.svmEgoSimulator = SvmEgoSimulator(matFileName)
class SvmEgoSimulatorTest(unittest.TestCase):
def setUp(self):
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
dataDir = PathDefaults.getDataDir() + "infoDiffusion/"
#dataDir = "/home/charanpal/Private/Postdoc/Code/APGL/data/"
matFileName = dataDir + "EgoAlterTransmissions.mat"
self.svmEgoSimulator = SvmEgoSimulator(matFileName)
def testSampleExamples(self):
self.svmEgoSimulator.sampleExamples(100)
self.assertEquals(self.svmEgoSimulator.examplesList.getNumSampledExamples(), 100)
def testModelSelection(self):
Cs = [1.0, 2.0]
kernel = "linear"
kernelParams = [0.0]
errorCosts = [0.1, 0.2]
folds = 5
sampleSize = 1000
CVal, kernelParamVal, errorCost, error = self.svmEgoSimulator.modelSelection(Cs, kernel, kernelParams, errorCosts, folds, sampleSize)
self.assertTrue(CVal in Cs)
self.assertTrue(kernelParamVal in kernelParams)
self.assertTrue(errorCost in errorCosts)
self.assertTrue(error >= 0.0 and error < 1.0)
def testEvaluateClassifier(self):
CVal = 1.0
kernel = "linear"
kernelParamVal = 0.0
errorCost = 0.5
folds = 6
sampleSize = 1000
invert = False
(means, vars) = self.svmEgoSimulator.evaluateClassifier(CVal, kernel, kernelParamVal, errorCost, folds, sampleSize, invert)
def testTrainClassifier(self):
CVal = 1.0
kernel = "linear"
kernelParamVal = 0.0
errorCost = 0.5
folds = 6
sampleSize = 1000
self.svmEgoSimulator.trainClassifier(CVal, kernel, kernelParamVal, errorCost, sampleSize)
def testGenerateRandomGraph(self):
egoFileName = PathDefaults.getDataDir() + "infoDiffusion/EgoData.csv"
alterFileName = PathDefaults.getDataDir() + "infoDiffusion/AlterData.csv"
numVertices = 1000
infoProb = 0.1
p = 0.1
neighbours = 10
generator = SmallWorldGenerator(p, neighbours)
graph = SparseGraph(VertexList(numVertices, 0))
graph = generator.generate(graph)
self.svmEgoSimulator.generateRandomGraph(egoFileName, alterFileName, infoProb, graph)
def testRunSimulation(self):
egoFileName = PathDefaults.getDataDir() + "infoDiffusion/EgoData.csv"
alterFileName = PathDefaults.getDataDir() + "infoDiffusion/AlterData.csv"
numVertices = 1000
infoProb = 0.1
p = 0.1
neighbours = 10
generator = SmallWorldGenerator(p, neighbours)
graph = SparseGraph(VertexList(numVertices, 0))
graph = generator.generate(graph)
CVal = 1.0
kernel = "linear"
kernelParamVal = 0.0
errorCost = 0.5
folds = 6
sampleSize = 1000
maxIterations = 5
self.svmEgoSimulator.trainClassifier(CVal, kernel, kernelParamVal, errorCost, sampleSize)
self.svmEgoSimulator.generateRandomGraph(egoFileName, alterFileName, infoProb, graph)
self.svmEgoSimulator.runSimulation(maxIterations)
import logging
import sys
import numpy
import random
from apgl.egograph.InfoExperiment import InfoExperiment
from apgl.egograph.SvmEgoSimulator import SvmEgoSimulator
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
numpy.random.seed(21)
random.seed(21)
examplesFileName = SvmInfoExperiment.getExamplesFileName()
folds = 5
cvSampleSize = SvmInfoExperiment.getNumCVExamples()
sampleSize = 75000
svmEgoSimulator = SvmEgoSimulator(examplesFileName)
classifier = svmEgoSimulator.getClassifier()
#classifier.setTermination(0.1)
C, kernel, kernelParamVal, errorCost = SvmInfoExperiment.loadSvmParams(SvmInfoExperiment.getSvmParamsFileName())
logging.info("Training SVM with C=" + str(C) + ", " + kernel + " kernel" + ", param=" + str(kernelParamVal) + ", sampleSize=" + str(sampleSize) + ", errorCost="+str(errorCost))
svmEgoSimulator.sampleExamples(cvSampleSize)
svmEgoSimulator.evaluateClassifier(C, kernel, kernelParamVal, errorCost, folds, sampleSize, True)
examplesFileName = SvmInfoExperiment.getExamplesFileName()
egoFileName = "../../data/EgoData.csv"
alterFileName = "../../data/AlterData.csv"
numVertices = SvmInfoExperiment.getNumVertices()
numVertices = 10000
infoProb = 0.1
graphTypes = ["SmallWorld", "ErdosRenyi"]
ps = [0.1, 0.003]
ks = [15, 15]
svmParamsFile = SvmInfoExperiment.getSvmParamsFileName()
sampleSize = SvmInfoExperiment.getNumSimulationExamples()
simulator = SvmEgoSimulator(examplesFileName)
CVal, kernel, kernelParamVal, errorCost = SvmInfoExperiment.loadSvmParams(svmParamsFile)
classifier = simulator.trainClassifier(CVal, kernel, kernelParamVal, errorCost, sampleSize)
preprocessor = simulator.getPreprocessor()
maxIterations = 5
eCsvReader = EgoCsvReader()
egoQuestionIds = eCsvReader.getEgoQuestionIds()
pajekWriter = PajekWriter()
simpleGraphWriter = SimpleGraphWriter()
vertexWriter = CsvVertexWriter()
for i in range(len(graphTypes)):
graphType = graphTypes[i]
import sys
import numpy
import random
from apgl.egograph.InfoExperiment import InfoExperiment
from apgl.egograph.SvmEgoSimulator import SvmEgoSimulator
from apgl.io.EgoCsvReader import EgoCsvReader
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
numpy.random.seed(21)
random.seed(21)
examplesFileName = SvmInfoExperiment.getExamplesFileName()
sampleSize = 86755
svmEgoSimulator = SvmEgoSimulator(examplesFileName)
preprocessor = svmEgoSimulator.getPreProcessor()
centerValues = preprocessor.getCentreVector()
svmParamsFileName = SvmInfoExperiment.getSvmParamsFileName() + "Linear.mat"
logging.info("Using SVM params from file " + svmParamsFileName)
C, kernel, kernelParamVal, errorCost = SvmInfoExperiment.loadSvmParams(svmParamsFileName)
svmEgoSimulator.trainClassifier(C, kernel, kernelParamVal, errorCost, sampleSize)
weights, b = svmEgoSimulator.getWeights()
numpy.set_printoptions(precision=3)
#Print the weights then their sorted values by indices and then value
sortedWeightsInds = numpy.flipud(numpy.argsort(abs(weights)))
k = 15 generator = SmallWorldGenerator(p, k) """ #A second set of parameters p = float(30)/numVertices generator = ErdosRenyiGenerator(p) simulationRepetitions = 5 maxIterations = 3 sampleSize = SvmInfoExperiment.getNumSimulationExamples() #sampleSize = 1000 svmParamsFile = SvmInfoExperiment.getSvmParamsFileName() CVal, kernel, kernelParamVal, errorCost = SvmInfoExperiment.loadSvmParams(svmParamsFile) simulator = SvmEgoSimulator(examplesFileName) simulator.trainClassifier(CVal, kernel, kernelParamVal, errorCost, sampleSize) egoCsvReader = EgoCsvReader() genderIndex = egoCsvReader.genderIndex ageIndex = egoCsvReader.ageIndex incomeIndex = egoCsvReader.incomeIndex townSizeIndex = egoCsvReader.townSizeIndex foodRiskIndex = egoCsvReader.foodRiskIndex experienceIndex = egoCsvReader.experienceIndex internetFreqIndex = egoCsvReader.internetFreqIndex peopleAtWorkIndex = egoCsvReader.peopleAtWorkIndex educationIndex = egoCsvReader.educationIndex professionIndices = numpy.zeros(egoCsvReader.numProfessions)