def loadRafFormat(self, DATASETNAME, test_split = 0.3): X = np.genfromtxt("datasets\\"+DATASETNAME+'.raf', delimiter=',')[:,:-1] ; X = [list(x) for x in X] Y = np.genfromtxt("datasets\\"+DATASETNAME+'.raf', delimiter=',', usecols=-1, dtype=str) X, Y = Util.shuffle_related_lists( X, Y ) nb_test = int ( len(Y) * test_split ) self.Ty = [ y for y in Y[:nb_test] ] self.Tx = [ list(x) for x in X[:nb_test] ] self.Y = [ y for y in Y[nb_test:] ] self.YY = self.Y[:] self.X = [ list(x) for x in X[nb_test:] ] self.X_transpose = [ list(v) for v in zip(*self.X) ] self.nb_data = len(self.X) self.nb_features = len(self.X[0]) self.features_name = [ "feature "+str(i) for i in range(self.nb_features) ] self.target_name = "target"
def loadFromSklearn(self, DATASETNAME, test_split = 0.3): # TODO add other datasets etc. if DATASETNAME == "digits": dataset = load_digits() if DATASETNAME == "iris": dataset = load_iris() # elif X, Y = Util.shuffle_related_lists( dataset["data"], dataset["target"] ) nb_test = int ( len(Y) * test_split ) self.Ty = [ y for y in Y[:nb_test] ] self.Tx = [ list(x) for x in X[:nb_test] ] self.Y = [ y for y in Y[nb_test:] ] self.YY = self.Y[:] self.X = [ list(x) for x in X[nb_test:] ] self.X_transpose = [ list(v) for v in zip(*self.X) ] self.nb_data = len(self.X) self.nb_features = len(self.X[0]) self.features_name = [ "feature "+str(i) for i in range(self.nb_features) ] self.target_name = "target"