[Proposed] df50-잘못함

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 = throw(fraudTrain,0.5)
df_results = try_1(df50, 0.5, 0.5, 10973.519989002007, 0.501)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.5, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.6, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.7, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.8, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.9, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.4, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.3, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 15000, 0.2, 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 0.891757 0.959609 0.817879 0.883094 0.925402 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10973.519989 0.501
1 GCN None 0.903969 0.959570 0.843420 0.897754 0.936666 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.500
2 GCN None 0.890647 0.960105 0.815103 0.881682 0.925188 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.600
3 GCN None 0.884541 0.960745 0.801777 0.874092 0.915326 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.700
4 GCN None 0.880377 0.950066 0.802887 0.870298 0.902739 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.800
5 GCN None 0.872884 0.959617 0.778456 0.859595 0.876013 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.900
6 GCN None 0.908132 0.956522 0.855081 0.902961 0.942754 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.400
7 GCN None 0.913961 0.952641 0.871183 0.910093 0.947405 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.300
8 GCN None 0.912295 0.962617 0.857857 0.907223 0.952030 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 15000.000000 0.200 
df_results = try_1(df50, 0.5, 0.5, 10000, 0.2)
df_results = try_1(df50, 0.5, 0.5, 10000, 0.2, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 10000, 0.18, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 10000, 0.16, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 10000, 0.14, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 10000, 0.12, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 10000, 0.10, prev_results=df_results)
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 0.909797 0.961827 0.853415 0.904384 0.946367 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10000 0.20
1 GCN None 0.910075 0.961274 0.854525 0.904762 0.946356 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10000 0.20
2 GCN None 0.910352 0.961298 0.855081 0.905084 0.946452 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10000 0.18
3 GCN None 0.913961 0.956522 0.867296 0.909726 0.947426 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10000 0.16
4 GCN None 0.912295 0.962041 0.858412 0.907277 0.948787 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10000 0.14
5 GCN None 0.915903 0.957265 0.870627 0.911893 0.950264 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10000 0.12
6 GCN None 0.917291 0.957395 0.873404 0.913473 0.951514 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 10000 0.10 
df_results = try_1(df50, 0.5, 0.5, 98000, 0.1)
df_results = try_1(df50, 0.5, 0.5, 96000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 94000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 92000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 90000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 88000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 86000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 84000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 82000, 0.1, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 80000, 0.1, prev_results=df_results)
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 0.964752 0.973416 0.955580 0.964416 0.987621 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 98000 0.1
1 GCN None 0.967527 0.965193 0.970017 0.967599 0.987662 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 96000 0.1
2 GCN None 0.967250 0.964660 0.970017 0.967331 0.987736 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 94000 0.1
3 GCN None 0.967250 0.964660 0.970017 0.967331 0.987737 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 92000 0.1
4 GCN None 0.968637 0.968368 0.968906 0.968637 0.987742 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 90000 0.1
5 GCN None 0.968360 0.967831 0.968906 0.968368 0.987859 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 88000 0.1
6 GCN None 0.966972 0.964128 0.970017 0.967063 0.987888 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 86000 0.1
7 GCN None 0.968082 0.970950 0.965019 0.967975 0.987927 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 84000 0.1
8 GCN None 0.967805 0.965727 0.970017 0.967867 0.987995 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 82000 0.1
9 GCN None 0.966972 0.964128 0.970017 0.967063 0.988069 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 80000 0.1 
df_results = try_1(df50, 0.5, 0.5, 88000, 0.2)
df_results = try_1(df50, 0.5, 0.5, 96000, 0.3, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 94000, 0.4, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 92000, 0.5, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 90000, 0.6, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 88000, 0.7, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 86000, 0.8, prev_results=df_results)
df_results = try_1(df50, 0.5, 0.5, 84000, 0.9, prev_results=df_results)
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 0.967250 0.972487 0.961688 0.967058 0.987065 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 88000 0.2
1 GCN None 0.965307 0.965039 0.965575 0.965307 0.986926 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 96000 0.3
2 GCN None 0.963364 0.963354 0.963354 0.963354 0.984985 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 94000 0.4
3 GCN None 0.937552 0.954965 0.918379 0.936315 0.968513 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 92000 0.5
4 GCN None 0.928948 0.950437 0.905053 0.927190 0.962901 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 90000 0.6
5 GCN None 0.913405 0.963863 0.858967 0.908397 0.957015 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 88000 0.7
6 GCN None 0.915071 0.953855 0.872293 0.911253 0.948241 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 86000 0.8
7 GCN None 0.895087 0.961713 0.822876 0.886894 0.931524 True Proposed 0.5 8409 amt 0.500059 3603 0.5 None 84000 0.9