maxLocalAuc.itemExpP = 1.0 maxLocalAuc.itemExpQ = 1.0 maxLocalAuc.lmbdas = numpy.linspace(0.5, 2.0, 7) maxLocalAuc.maxIterations = 100 maxLocalAuc.metric = "f1" maxLocalAuc.normalise = True maxLocalAuc.numAucSamples = 10 #maxLocalAuc.numProcesses = 1 maxLocalAuc.numRecordAucSamples = 100 maxLocalAuc.numRowSamples = 30 maxLocalAuc.rate = "optimal" maxLocalAuc.recommendSize = 5 maxLocalAuc.recordStep = 1 maxLocalAuc.rho = 1.0 maxLocalAuc.t0 = 1.0 maxLocalAuc.t0s = 2.0**-numpy.arange(-1, 6, 1) maxLocalAuc.validationSize = 5 maxLocalAuc.validationUsers = 0 t0s = 2.0**-numpy.arange(-2, 8, 1) #maxLocalAuc.maxIterations = 1 #maxLocalAuc.numProcesses = 1 meanObjsList = [] stdObjsList = [] initialAlgs = ["rand", "svd"] rates = ["optimal", "constant"] if saveResults:
maxLocalAuc.itemExpP = 0.0 maxLocalAuc.itemExpQ = 0.0 maxLocalAuc.ks = numpy.array([k2]) maxLocalAuc.lmbdas = numpy.linspace(0.5, 2.0, 7) maxLocalAuc.maxIterations = 500 maxLocalAuc.metric = "f1" maxLocalAuc.normalise = True maxLocalAuc.numAucSamples = 10 maxLocalAuc.numProcesses = 1 maxLocalAuc.numRecordAucSamples = 100 maxLocalAuc.numRowSamples = 30 maxLocalAuc.rate = "constant" maxLocalAuc.recordStep = 10 maxLocalAuc.rho = 1.0 maxLocalAuc.t0 = 1.0 maxLocalAuc.t0s = 2.0**-numpy.arange(7, 12, 1) maxLocalAuc.validationSize = 3 maxLocalAuc.validationUsers = 0 os.system('taskset -p 0xffffffff %d' % os.getpid()) logging.debug("Starting training") losses = [("tanh", 0.25), ("tanh", 0.5), ("tanh", 1.0), ("tanh", 2.0), ("hinge", 1), ("square", 1), ("logistic", 0.5), ("logistic", 1.0), ("logistic", 2.0), ("sigmoid", 0.5), ("sigmoid", 1.0), ("sigmoid", 2.0)] def computeTestAuc(args): trainX, testX, maxLocalAuc, U, V = args numpy.random.seed(21) logging.debug(maxLocalAuc) #maxLocalAuc.learningRateSelect(trainX) U, V, trainMeasures, testMeasures, iterations, time = maxLocalAuc.learnModel(trainX, U=U, V=V, verbose=True)
maxLocalAuc.loss = "hinge" maxLocalAuc.maxIterations = 500 maxLocalAuc.maxNorm = 100 maxLocalAuc.metric = "f1" maxLocalAuc.normalise = False maxLocalAuc.numAucSamples = 10 maxLocalAuc.numProcesses = multiprocessing.cpu_count() maxLocalAuc.numRecordAucSamples = 200 maxLocalAuc.numRowSamples = 15 maxLocalAuc.rate = "optimal" maxLocalAuc.recordStep = 10 maxLocalAuc.reg = False maxLocalAuc.rho = 1.0 maxLocalAuc.startAverage = 100 maxLocalAuc.t0 = 1.0 maxLocalAuc.t0s = 2.0**-numpy.arange(1, 12, 2) maxLocalAuc.validationSize = 5 maxLocalAuc.validationUsers = 0.0 if saveResults: X = DatasetUtils.getDataset(dataset, nnz=100000) print(X.shape, X.nnz) print(maxLocalAuc) maxLocalAuc.lmbdaU = 0.25 maxLocalAuc.lmbdaV = 0.25 meanObjs1, paramDict = maxLocalAuc.learningRateSelect(X) maxLocalAuc.lmbdaU = 0.03125 maxLocalAuc.lmbdaV = 0.25 meanObjs2, paramDict = maxLocalAuc.learningRateSelect(X)