def __init__(self): self.data = Data()
import pandas as pd from sklearn.model_selection import train_test_split from sklearn.svm import OneClassSVM from src.common.my_data import Data data = Data() train_agg_flg_0 = data.output.train_agg_flg_0 train_agg_flg_1 = data.output.train_agg_flg_1 train = pd.read_table(train_agg_flg_0).drop('USRID', axis=1) test = pd.read_table(train_agg_flg_1).drop('USRID', axis=1) train_noID = train[0:70000] test_noID = train[70000:72000].append(test[0:100]) svm = OneClassSVM() svm.fit(train_noID) result = svm.predict(test_noID) result0 = result[0:2000] result1 = result[2000:] n_result0_0 = 0 n_result1_1 = 0 n = 0 for i in result0: if i == 1: n_result0_0+=1 n+=1