def test_CNN(x, y): print('Condensed Nearest Neighbour') CNN = CondensedNearestNeighbour(verbose=verbose) cnnx, cnny = CNN.fit_transform(x, y) print('One-Sided Selection') OSS = OneSidedSelection(verbose=verbose) ossx, ossy = OSS.fit_transform(x, y) print('BalanceCascade') BS = BalanceCascade(verbose=verbose) bsx, bsy = BS.fit_transform(x, y)
def balance_data_ensemblesampling_balance_cascade(self): ''' Balance data using balance cascade. ''' x = self.X y = self.y y.shape = (len(self.y)) verbose = True BS = BalanceCascade(verbose=verbose) bsx, bsy = BS.fit_transform(x, y) self.X = bsx self.y = bsy self.y.shape = (len(self.y), 1)
def test_CNN(x, y): print('Condensed Nearest Neighbour') CNN = CondensedNearestNeighbour(indices_support=indices_support, verbose=verbose) cnnx, cnny, idx_tmp = CNN.fit_transform(x, y) print ('Indices selected') print(idx_tmp) print('One-Sided Selection') OSS = OneSidedSelection(indices_support=indices_support, verbose=verbose) ossx, ossy, idx_tmp = OSS.fit_transform(x, y) print ('Indices selected') print(idx_tmp) print('BalanceCascade') BS = BalanceCascade(verbose=verbose) bsx, bsy = BS.fit_transform(x, y)
def test_CNN(x, y,c=0,ratio='auto'): if(c==0): print('Condensed Nearest Neighbour') CNN = CondensedNearestNeighbour(indices_support=indices_support, verbose=verbose) x,y, idx_tmp = CNN.fit_transform(x, y) print ('Indices selected') print(idx_tmp) elif(c==1): print('One-Sided Selection') OSS = OneSidedSelection(indices_support=indices_support, verbose=verbose) x,y, idx_tmp = OSS.fit_transform(x, y) print ('Indices selected') print(idx_tmp) elif(c==2): print('BalanceCascade') BS = BalanceCascade(ratio='auto',verbose=verbose) x,y = BS.fit_transform(x, y) return x,y