def poisciParametre(self, X,Y): activation_functions = ['multiquadric' , 'softlim', 'inv_multiquadric', 'gaussian', 'tanh', 'sine', 'tribas', 'inv_tribas', 'sigmoid'] n_hiddens = [50, 100,200, 500,800, 900, 1000, 1500, 2000, 3000, 5000, 10000]#3, 30,50,100] parameters = [] alphas = [1.0,0.7,0.5,0.0]#0.0,0.2,0.4,0.5,0.7,0.9,1.0] nrOfTrials = len(activation_functions)*len(alphas) * len(n_hiddens) trial = 1 np.random.seed(np.random.randint(10000000)) for n_hidden in n_hiddens: for alpha in alphas: for actFunction in activation_functions: cls = GenELMClassifier(hidden_layer = RandomLayer(n_hidden = n_hidden, activation_func = actFunction, alpha=alpha)) parameter = Helpers.cv(X,Y,cls,5, printing = False) parameter = parameter+ [n_hidden, alpha, actFunction, "normal"] parameters.append(parameter) print(parameter, "%d/%d" %(trial,nrOfTrials)) Helpers.pickleListAppend2(parameter, "parametersELM.p") # parameter = Helpers.cv(X,Y,BaggingClassifier(cls,n_estimators=30),10, printing = False) # parameter = parameter+ [n_hidden, alpha, actFunction, "bagged"] # parameters.append(parameter) # print(parameter, "%d/%d" %(trial,nrOfTrials)) trial = trial+1 # pickle.dump(parameters,open("parametersMultiQuadric.p","wb")) return
def main3(): #Testira nekaj algoritmov X, Y = readData(trainFtrFile="data/trainFtrExtended_200f.csv", trainClsFile= "data/trainClsExtended.csv", deleteFirstNFeatures=2) # X, Y = readData(1000) elmc = GenELMClassifier(hidden_layer = RandomLayer(n_hidden = 50, activation_func = 'multiquadric', alpha=1.0)) baggedElmc = BaggingClassifier(elmc, n_estimators=10) ada = AdaBoostClassifier(n_estimators=30) rf = sklearn.ensemble.RandomForestClassifier(n_estimators = 50, n_jobs = 1) np.random.seed(np.random.randint(10000000)) adaRf = AdaBoostClassifier(rf, n_estimators=20) print ("elmc: ", Helpers.cv(X,Y,elmc)) print ("baggedElmc: ", Helpers.cv(X,Y,baggedElmc)) print ("adaTree: ", Helpers.cv(X,Y,ada)) print ("rf: ", Helpers.cv(X,Y,rf)) print ("adaRf: ", Helpers.cv(X,Y,adaRf))
def main5(): #Cross validation 3 algoritmov X, Y = readData() tree = DecisionTreeClassifier() elmc = GenELMClassifier(hidden_layer = RandomLayer(n_hidden = 100, activation_func = 'multiquadric', alpha=1.0)) baggedelmc = Bagging(elmc, n_estimators=20,ratioCutOffEstimators=0) print ("baggedelmc: ", Helpers.cv(X,Y,baggedelmc))
def main1(): X, Y = readData(10000) elmc = ELMClassifier(n_hidden=100, activation_func='gaussian') baggedElmc = BaggingClassifier(elmc) #baggedElmc.fit(X,Y) rhl = RandomLayer(n_hidden=500, activation_func='gaussian') genElmc = GenELMClassifier(rhl) tr, ts, trRaw, tsRaw, prec, precTress, duration = Helpers.cv(X,Y,baggedElmc,3, printing=True) plt.scatter(tr, ts, alpha=0.5, marker='D', c='r') plt.scatter(trRaw, tsRaw, alpha=0.5, marker='D', c='b') plt.show() plt.scatter(prec, precTress,alpha=0.9, marker='D', c='r') plt.show()