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)

df_results = try_6(fraudTrain, 0.2,1e7,0.96)
df_results = try_6(fraudTrain, 0.2,1e7,0.94, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.92, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.9, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.88, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.86, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.84, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.82, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.80, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.78, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.76, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.74, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.72, prev_results=df_results)
df_results = try_6(fraudTrain, 0.2,1e7,0.7, 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	0.969499	0.908387	0.941806	0.924795	0.989259	True	Proposed	0.2	22522	amt	0.200293	7508	0.199121	None	10000000.0	0.96
1	GCN	None	0.965503	0.901242	0.935525	0.918064	0.987353	True	Proposed	0.2	22522	amt	0.197807	7508	0.206580	None	10000000.0	0.94
2	GCN	None	0.955115	0.888591	0.885619	0.887102	0.984401	True	Proposed	0.2	22522	amt	0.200293	7508	0.199121	None	10000000.0	0.92
3	GCN	None	0.958977	0.874307	0.931714	0.902098	0.983667	True	Proposed	0.2	22522	amt	0.199050	7508	0.202850	None	10000000.0	0.90
4	GCN	None	0.949254	0.850236	0.919057	0.883308	0.980431	True	Proposed	0.2	22522	amt	0.197007	7508	0.208977	None	10000000.0	0.88
5	GCN	None	0.948055	0.821918	0.938137	0.876190	0.979554	True	Proposed	0.2	22522	amt	0.201359	7508	0.195924	None	10000000.0	0.86
6	GCN	None	0.948189	0.826012	0.941991	0.880197	0.981559	True	Proposed	0.2	22522	amt	0.199316	7508	0.202051	None	10000000.0	0.84
7	GCN	None	0.928476	0.813968	0.840105	0.826830	0.972653	True	Proposed	0.2	22522	amt	0.198917	7508	0.203250	None	10000000.0	0.82
8	GCN	None	0.936867	0.810303	0.892523	0.849428	0.976111	True	Proposed	0.2	22522	amt	0.200160	7508	0.199521	None	10000000.0	0.80
9	GCN	None	0.937400	0.805353	0.898236	0.849262	0.974586	True	Proposed	0.2	22522	amt	0.201225	7508	0.196324	None	10000000.0	0.78
10	GCN	None	0.937134	0.793696	0.925184	0.854411	0.974375	True	Proposed	0.2	22522	amt	0.200204	7508	0.199387	None	10000000.0	0.76
11	GCN	None	0.923015	0.799333	0.812331	0.805780	0.968231	True	Proposed	0.2	22522	amt	0.201137	7508	0.196590	None	10000000.0	0.74
12	GCN	None	0.912360	0.774086	0.785570	0.779786	0.961307	True	Proposed	0.2	22522	amt	0.200826	7508	0.197523	None	10000000.0	0.72
13	GCN	None	0.904635	0.780702	0.747416	0.763696	0.962761	True	Proposed	0.2	22522	amt	0.197940	7508	0.206180	None	10000000.0	0.70

df_results = try_6(fraudTrain, 0.009,1e7,0.8)
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