[Proposed] df50tr, df005ts_함수

Author

김보람

Published

February 14, 2024

imports

import pandas as pd
import numpy as np
import sklearn
import pickle 
import time 
import datetime
import warnings
warnings.filterwarnings('ignore')
%run ../function_proposed_gcn.py
with open('../fraudTrain.pkl', 'rb') as file:
    fraudTrain = pickle.load(file)

df50_train, df[::10]: test

df_results = try_4(fraudTrain, 10, 10973.519989002007, 0.501)

ymdhms = datetime.datetime.fromtimestamp(time.time()).strftime('%Y%m%d-%H%M%S') 
df_results.to_csv(f'./results/{ymdhms}-proposed.csv',index=False)
df_results
model time acc pre rec f1 auc graph_based method throw_rate train_size train_cols train_frate test_size test_frate hyper_params theta gamma
0 GCN None 0.958344 0.10118 0.821918 0.18018 0.911725 True Proposed 0.131927 9009 amt 0.499611 26215 0.005569 None 10973.519989 0.501 
df_results
model time acc pre rec f1 auc graph_based method throw_rate train_size train_cols train_frate test_size test_frate hyper_params theta gamma
0 GCN None 0.958344 0.101180 0.821918 0.180180 0.911725 True Proposed 0.131927 9009 amt 0.499611 26215 0.005569 None 10973.519989 0.501
1 GCN None 0.956971 0.104098 0.783951 0.183792 0.889444 True Proposed 0.132637 9009 amt 0.500611 26215 0.006180 None 10973.519989 0.600
2 GCN None 0.966088 0.103556 0.755725 0.182153 0.866406 True Proposed 0.131672 9009 amt 0.500278 26215 0.004997 None 10973.519989 0.700
3 GCN None 0.953805 0.087529 0.773973 0.157272 0.859619 True Proposed 0.131530 9009 amt 0.498057 26215 0.005569 None 10973.519989 0.800
4 GCN None 0.950067 0.092659 0.783133 0.165711 0.860542 True Proposed 0.132211 9009 amt 0.498501 26215 0.006332 None 10973.519989 0.900 
df_results = try_4(fraudTrain, 10, 10973, 0.1)
df_results = try_4(fraudTrain, 10, 10973, 0.2, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 10973, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 10973, 0.4, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 10973, 0.5, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 10973, 0.6, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 10973, 0.7, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 10973, 0.8, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 10973, 0.9, prev_results=df_results)

dhms = datetime.datetime.fromtimestamp(time.time()).strftime('%Y%m%d-%H%M%S') 
df_results.to_csv(f'../results/{ymdhms}-proposed.csv',index=False)

df_results
NameError: name 'ymdhms' is not defined

ymdhms = datetime.datetime.fromtimestamp(time.time()).strftime('%Y%m%d-%H%M%S') 
df_results.to_csv(f'../results/{ymdhms}-proposed.csv',index=False)

df_results
model time acc pre rec f1 auc graph_based method throw_rate train_size train_cols train_frate test_size test_frate hyper_params theta gamma
0 GCN None 0.953805 0.106222 0.809249 0.187793 0.931387 True Proposed 0.133063 9009 amt 0.501055 26215 0.006599 None 10973 0.1
1 GCN None 0.966012 0.137275 0.820359 0.235193 0.926498 True Proposed 0.132750 9009 amt 0.500500 26215 0.006370 None 10973 0.2
2 GCN None 0.960633 0.107365 0.823129 0.189953 0.907009 True Proposed 0.131785 9009 amt 0.498945 26215 0.005607 None 10973 0.3
3 GCN None 0.964372 0.118744 0.770701 0.205782 0.895444 True Proposed 0.132097 9009 amt 0.499056 26215 0.005989 None 10973 0.4
4 GCN None 0.953996 0.096228 0.796178 0.171703 0.885994 True Proposed 0.133176 9009 amt 0.503275 26215 0.005989 None 10973 0.5
5 GCN None 0.945833 0.080795 0.792208 0.146635 0.885693 True Proposed 0.133063 9009 amt 0.503164 26215 0.005874 None 10973 0.6
6 GCN None 0.955979 0.095857 0.742138 0.169784 0.865008 True Proposed 0.132949 9009 amt 0.502165 26215 0.006065 None 10973 0.7
7 GCN None 0.957353 0.097908 0.745223 0.173077 0.864288 True Proposed 0.132438 9009 amt 0.500389 26215 0.005989 None 10973 0.8
8 GCN None 0.949189 0.085271 0.780645 0.153748 0.848454 True Proposed 0.132268 9009 amt 0.499944 26215 0.005913 None 10973 0.9 
df_results = try_4(fraudTrain, 10, 10000, 0.3)
df_results = try_4(fraudTrain, 10, 9800, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9400, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9200, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9000, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 8800, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 8600, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 8400, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 8200, 0.3, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 8000, 0.3, prev_results=df_results)

ymdhms = datetime.datetime.fromtimestamp(time.time()).strftime('%Y%m%d-%H%M%S') 
df_results.to_csv(f'../results/{ymdhms}-proposed.csv',index=False)

df_results
model time acc pre rec f1 auc graph_based method throw_rate train_size train_cols train_frate test_size test_frate hyper_params theta gamma
0 GCN None 0.957047 0.095789 0.789116 0.170839 0.898871 True Proposed 0.132438 9009 amt 0.501499 26215 0.005607 None 10000 0.3
1 GCN None 0.963151 0.115238 0.765823 0.200331 0.918913 True Proposed 0.132665 9009 amt 0.501166 26215 0.006027 None 9800 0.3
2 GCN None 0.953538 0.097212 0.865772 0.174797 0.925368 True Proposed 0.131161 9009 amt 0.496281 26215 0.005684 None 9600 0.3
3 GCN None 0.955484 0.095770 0.779221 0.170576 0.891125 True Proposed 0.133375 9009 amt 0.504385 26215 0.005874 None 9400 0.3
4 GCN None 0.962731 0.121771 0.840764 0.212732 0.919305 True Proposed 0.132268 9009 amt 0.499722 26215 0.005989 None 9200 0.3
5 GCN None 0.938051 0.075014 0.860927 0.138004 0.917114 True Proposed 0.133034 9009 amt 0.503386 26215 0.005760 None 9000 0.3
6 GCN None 0.958497 0.087674 0.731884 0.156589 0.859628 True Proposed 0.132239 9009 amt 0.501721 26215 0.005264 None 8800 0.3
7 GCN None 0.961015 0.106460 0.745223 0.186306 0.891528 True Proposed 0.132523 9009 amt 0.500722 26215 0.005989 None 8600 0.3
8 GCN None 0.955369 0.089457 0.788732 0.160689 0.913984 True Proposed 0.133034 9009 amt 0.504385 26215 0.005417 None 8400 0.3
9 GCN None 0.949723 0.076703 0.781022 0.139687 0.899420 True Proposed 0.132154 9009 amt 0.501499 26215 0.005226 None 8200 0.3
10 GCN None 0.947168 0.078338 0.782313 0.142415 0.885496 True Proposed 0.131019 9009 amt 0.495948 26215 0.005607 None 8000 0.3 
df_results = try_4(fraudTrain, 10, 9600, 0.2)
df_results = try_4(fraudTrain, 10, 9600, 0.18, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.16, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.14, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.12, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.1, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.08, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.06, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.04, prev_results=df_results)
df_results = try_4(fraudTrain, 10, 9600, 0.02, prev_results=df_results)

ymdhms = datetime.datetime.fromtimestamp(time.time()).strftime('%Y%m%d-%H%M%S') 
df_results.to_csv(f'../results/{ymdhms}-proposed.csv',index=False)

df_results
model time acc pre rec f1 auc graph_based method throw_rate train_size train_cols train_frate test_size test_frate hyper_params theta gamma
0 GCN None 0.961663 0.114621 0.841060 0.201747 0.932483 True Proposed 0.132012 9009 amt 0.499389 26215 0.005760 None 9600 0.20
1 GCN None 0.959565 0.117140 0.795322 0.204204 0.911232 True Proposed 0.133261 9009 amt 0.502054 26215 0.006523 None 9600 0.18
2 GCN None 0.949304 0.087018 0.815789 0.157261 0.926008 True Proposed 0.131586 9009 amt 0.497613 26215 0.005798 None 9600 0.16
3 GCN None 0.950906 0.085199 0.855072 0.154957 0.939830 True Proposed 0.131189 9009 amt 0.497613 26215 0.005264 None 9600 0.14
4 GCN None 0.961434 0.115108 0.825806 0.202052 0.919188 True Proposed 0.131927 9009 amt 0.498612 26215 0.005913 None 9600 0.12
5 GCN None 0.963418 0.115163 0.764331 0.200167 0.911616 True Proposed 0.132239 9009 amt 0.499611 26215 0.005989 None 9600 0.10
6 GCN None 0.964944 0.132425 0.829268 0.228380 0.939729 True Proposed 0.133489 9009 amt 0.503719 26215 0.006256 None 9600 0.08
7 GCN None 0.953347 0.087356 0.780822 0.157133 0.921464 True Proposed 0.131189 9009 amt 0.496725 26215 0.005569 None 9600 0.06
8 GCN None 0.949456 0.088090 0.801282 0.158730 0.927804 True Proposed 0.132921 9009 amt 0.502387 26215 0.005951 None 9600 0.04
9 GCN None 0.956857 0.105777 0.802469 0.186916 0.921032 True Proposed 0.133630 9009 amt 0.504496 26215 0.006180 None 9600 0.02