[FRAUD] 데이터정리 시도(GCN_X범주)

imports

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt 
import networkx as nx
import sklearn
import xgboost as xgb

# sklearn
from sklearn import model_selection # split함수이용
from sklearn import ensemble # RF,GBM
from sklearn import metrics
from sklearn.metrics import precision_score, recall_score, f1_score
from sklearn.svm import SVC
from sklearn.ensemble import RandomForestClassifier
from sklearn.naive_bayes import GaussianNB

# gnn
import torch
import torch.nn.functional as F
import torch_geometric
from torch_geometric.nn import GCNConv

/home/coco/anaconda3/envs/py38/lib/python3.8/site-packages/torch_geometric/typing.py:18: UserWarning: An issue occurred while importing 'pyg-lib'. Disabling its usage. Stacktrace: /home/coco/anaconda3/envs/py38/lib/python3.8/site-packages/libpyg.so: undefined symbol: _ZN2at4_ops12split_Tensor4callERKNS_6TensorEN3c106SymIntEl
  warnings.warn(f"An issue occurred while importing 'pyg-lib'. "
/home/coco/anaconda3/envs/py38/lib/python3.8/site-packages/torch_geometric/typing.py:31: UserWarning: An issue occurred while importing 'torch-scatter'. Disabling its usage. Stacktrace: /home/coco/anaconda3/envs/py38/lib/python3.8/site-packages/torch_scatter/_scatter_cuda.so: undefined symbol: _ZNK3c107SymBool10guard_boolEPKcl
  warnings.warn(f"An issue occurred while importing 'torch-scatter'. "
/home/coco/anaconda3/envs/py38/lib/python3.8/site-packages/torch_geometric/typing.py:42: UserWarning: An issue occurred while importing 'torch-sparse'. Disabling its usage. Stacktrace: /home/coco/anaconda3/envs/py38/lib/python3.8/site-packages/torch_sparse/_diag_cuda.so: undefined symbol: _ZN3c106detail19maybe_wrap_dim_slowIlEET_S2_S2_b
  warnings.warn(f"An issue occurred while importing 'torch-sparse'. "

def down_sample_textbook(df):
    df_majority = df[df.is_fraud==0].copy()
    df_minority = df[df.is_fraud==1].copy()
    df_maj_dowsampled = sklearn.utils.resample(df_majority, n_samples=len(df_minority), replace=False, random_state=42)
    df_downsampled = pd.concat([df_minority, df_maj_dowsampled])
    return df_downsampled

def compute_time_difference(group):
    n = len(group)
    result = []
    for i in range(n):
        for j in range(n):
            time_difference = abs(group.iloc[i].trans_date_trans_time.value - group.iloc[j].trans_date_trans_time.value)
            result.append([group.iloc[i].name, group.iloc[j].name, time_difference])
    return result

def mask(df):
    df_tr,df_test = sklearn.model_selection.train_test_split(df, random_state=42)
    N = len(df)
    train_mask = [i in df_tr.index for i in range(N)]
    test_mask = [i in df_test.index for i in range(N)]
    train_mask = np.array(train_mask)
    test_mask = np.array(test_mask)
    return train_mask, test_mask

def edge_index_selected(edge_index):
    theta = edge_index[:,2].mean()
    edge_index[:,2] = (np.exp(-edge_index[:,2]/theta) != 1)*(np.exp(-edge_index[:,2]/theta))
    edge_index = edge_index.tolist()
    mean_ = np.array(edge_index)[:,2].mean()
    selected_edges = [(int(row[0]), int(row[1])) for row in edge_index if row[2] > mean_]
    edge_index_selected = torch.tensor(selected_edges, dtype=torch.long).t()
    return edge_index_selected

fraudTrain = pd.read_csv("~/Desktop/fraudTrain.csv").iloc[:,1:]

fraudTrain = fraudTrain.assign(trans_date_trans_time= list(map(lambda x: pd.to_datetime(x), fraudTrain.trans_date_trans_time)))
fraudTrain

	trans_date_trans_time	cc_num	merchant	category	amt	first	last	gender	street	city	...	lat	long	city_pop	job	dob	trans_num	unix_time	merch_lat	merch_long	is_fraud
0	2019-01-01 00:00:00	2.703190e+15	fraud_Rippin, Kub and Mann	misc_net	4.97	Jennifer	Banks	F	561 Perry Cove	Moravian Falls	...	36.0788	-81.1781	3495	Psychologist, counselling	1988-03-09	0b242abb623afc578575680df30655b9	1325376018	36.011293	-82.048315	0
1	2019-01-01 00:00:00	6.304230e+11	fraud_Heller, Gutmann and Zieme	grocery_pos	107.23	Stephanie	Gill	F	43039 Riley Greens Suite 393	Orient	...	48.8878	-118.2105	149	Special educational needs teacher	1978-06-21	1f76529f8574734946361c461b024d99	1325376044	49.159047	-118.186462	0
2	2019-01-01 00:00:00	3.885950e+13	fraud_Lind-Buckridge	entertainment	220.11	Edward	Sanchez	M	594 White Dale Suite 530	Malad City	...	42.1808	-112.2620	4154	Nature conservation officer	1962-01-19	a1a22d70485983eac12b5b88dad1cf95	1325376051	43.150704	-112.154481	0
3	2019-01-01 00:01:00	3.534090e+15	fraud_Kutch, Hermiston and Farrell	gas_transport	45.00	Jeremy	White	M	9443 Cynthia Court Apt. 038	Boulder	...	46.2306	-112.1138	1939	Patent attorney	1967-01-12	6b849c168bdad6f867558c3793159a81	1325376076	47.034331	-112.561071	0
4	2019-01-01 00:03:00	3.755340e+14	fraud_Keeling-Crist	misc_pos	41.96	Tyler	Garcia	M	408 Bradley Rest	Doe Hill	...	38.4207	-79.4629	99	Dance movement psychotherapist	1986-03-28	a41d7549acf90789359a9aa5346dcb46	1325376186	38.674999	-78.632459	0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1048570	2020-03-10 16:07:00	6.011980e+15	fraud_Fadel Inc	health_fitness	77.00	Haley	Wagner	F	05561 Farrell Crescent	Annapolis	...	39.0305	-76.5515	92106	Accountant, chartered certified	1943-05-28	45ecd198c65e81e597db22e8d2ef7361	1362931649	38.779464	-76.317042	0
1048571	2020-03-10 16:07:00	4.839040e+15	fraud_Cremin, Hamill and Reichel	misc_pos	116.94	Meredith	Campbell	F	043 Hanson Turnpike	Hedrick	...	41.1826	-92.3097	1583	Geochemist	1999-06-28	c00ce51c6ebb7657474a77b9e0b51f34	1362931670	41.400318	-92.726724	0
1048572	2020-03-10 16:08:00	5.718440e+11	fraud_O'Connell, Botsford and Hand	home	21.27	Susan	Mills	F	005 Cody Estates	Louisville	...	38.2507	-85.7476	736284	Engineering geologist	1952-04-02	17c9dc8b2a6449ca2473726346e58e6c	1362931711	37.293339	-84.798122	0
1048573	2020-03-10 16:08:00	4.646850e+18	fraud_Thompson-Gleason	health_fitness	9.52	Julia	Bell	F	576 House Crossroad	West Sayville	...	40.7320	-73.1000	4056	Film/video editor	1990-06-25	5ca650881b48a6a38754f841c23b77ab	1362931718	39.773077	-72.213209	0
1048574	2020-03-10 16:08:00	2.283740e+15	fraud_Buckridge PLC	misc_pos	6.81	Shannon	Williams	F	9345 Spencer Junctions Suite 183	Alpharetta	...	34.0770	-84.3033	165556	Prison officer	1997-12-27	8d0a575fe635bbde12f1a2bffc126731	1362931730	33.601468	-83.891921	0

1048575 rows × 22 columns

데이터정리

_df1 = fraudTrain[fraudTrain["is_fraud"] == 0].sample(frac=0.20, random_state=42)
_df2 = fraudTrain[fraudTrain["is_fraud"] == 1]
df02 = pd.concat([_df1,_df2])
df02.shape

(214520, 22)

df50 = down_sample_textbook(df02)
df50 = df50.reset_index()
df50.shape

(12012, 23)

---

tr/test

mask(df50)

(array([False,  True,  True, ...,  True, False,  True]),
 array([ True, False, False, ..., False,  True, False]))

train_mask, test_mask = mask(df50)

---

edge_index 설정

# groups = df50.groupby('cc_num')
# edge_index_list_plus = [compute_time_difference(group) for _, group in groups]
# edge_index_list_plus_flat = [item for sublist in edge_index_list_plus for item in sublist]
# edge_index_list_plus_nparr = np.array(edge_index_list_plus_flat)
# np.save('edge_index_list_plus50.npy', edge_index_list_plus_nparr)

edge_index = np.load('edge_index_list_plus50.npy').astype(np.float64)
edge_index.shape

(200706, 3)

edge_index_selected = edge_index_selected(edge_index)

---

data설정(x, edge_index, y)

- x에 시간을 추가해 보자.

---

정리

구분	Train	Test	모형	변경사항	비고
분석1	df50_tr	df50_test	GCN1	x:amt	기본
분석2	df50_tr	df50_test	GCN1	x: amt, time	처음 돌리면 계속 f1값이 0나옴.. 2번 돌려야 되는데.흠
분석3	df50_tr	df50_test	GCN1	x:amt, category	category수치형변환
분석4	df50_tr	df50_test	GCN1	x: amt, distance
분석5	df50_tr	df50_test	GCN1	x: amt, time, category
분석6	df50_tr	df50_test	GCN1	x: amt, time, distance
분석7	df50_tr	df50_test	GCN1	x: amt, category, distance
분석8	df50_tr	df50_test	GCN1	x: amt, category, category, distacne

- result2 값 비교용

• 전

lst = [_results1, _results2,_results3,_results4,_results5, _results6, _results7, _results8]
pd.concat(lst)

	accuracy_score	precision_score	recall_score	f1_score
분석1	0.902098	0.862478	0.959130	0.908240
분석2	0.494838	0.000000	0.000000	0.000000
분석3	0.899767	0.861905	0.954515	0.905849
분석4	0.913753	0.866978	0.979565	0.919839
분석5	0.906094	0.866904	0.961767	0.911875
분석6	0.913753	0.866978	0.979565	0.919839
분석7	0.914752	0.868498	0.979565	0.920694
분석8	0.913420	0.861415	0.987475	0.920147

• 후

lst = [_results1, _results2,_results3,_results4,_results5, _results6, _results7, _results8]
pd.concat(lst)

	accuracy_score	precision_score	recall_score	f1_score
분석1	0.902098	0.862478	0.959130	0.908240
분석2	0.897436	0.855798	0.958471	0.904229
분석3	0.899767	0.861905	0.954515	0.905849
분석4	0.913753	0.866978	0.979565	0.919839
분석5	0.906094	0.866904	0.961767	0.911875
분석6	0.913753	0.866978	0.979565	0.919839
분석7	0.914752	0.868498	0.979565	0.920694
분석8	0.913420	0.861415	0.987475	0.920147

---

분석 1(GCN): amt

x = torch.tensor(df50['amt'], dtype=torch.float).reshape(-1,1)
y = torch.tensor(df50['is_fraud'],dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index = edge_index_selected, y=y, train_mask = train_mask, test_mask = test_mask)
data



torch.manual_seed(202250926)
class GCN2(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(1, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN2()
optimizer = torch.optim.Adam(model.parameters(), lr=0.01, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()
model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results1= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석1'])
_results1

	accuracy_score	precision_score	recall_score	f1_score
분석1	0.902098	0.862478	0.95913	0.90824

---

b,W = list(model.conv1.parameters())

b,W

(Parameter containing:
 tensor([ 1.0399e+00, -2.5585e-01,  1.0566e-09, -4.7506e-03, -6.4289e-03,
         -3.0055e-02,  3.6866e-01,  5.0163e-01,  1.0861e+00, -4.2169e-02,
          2.1598e-07,  3.0382e-01, -1.5149e-01, -2.7623e-01,  6.8350e-01,
         -1.5861e-03, -7.9415e-01, -7.1081e-03, -7.7902e-03,  6.8695e-01,
         -1.5212e-01, -9.4722e-05, -8.9782e-01,  7.6822e-01, -5.0592e-02,
          8.1692e-01, -5.2181e-02, -2.1817e-02, -9.1607e-01,  7.9294e-01,
          1.0454e-09, -8.1069e-06], requires_grad=True),
 Parameter containing:
 tensor([[-4.8377e-03,  1.1379e-10],
         [-1.4998e-01,  2.4131e-10],
         [-2.1920e-01, -1.5126e-11],
         [-2.2593e-01,  5.3614e-11],
         [-3.0312e-01, -3.9329e-11],
         [-1.4487e-01,  5.2859e-11],
         [-4.1131e-02,  2.3188e-10],
         [ 1.1339e-01,  2.6153e-11],
         [-5.1434e-03,  1.2029e-10],
         [-2.4512e-01,  4.0292e-11],
         [-3.1688e-01,  4.6636e-11],
         [ 5.1427e-02,  2.0959e-10],
         [-2.6159e-01,  3.8738e-11],
         [-2.7135e-02, -1.3194e-11],
         [ 4.2538e-03,  1.4590e-10],
         [-2.1030e-01, -1.7922e-10],
         [ 2.4350e-02,  2.4147e-11],
         [-2.5375e-01, -5.6721e-11],
         [-2.4148e-01,  1.2034e-10],
         [ 2.5918e-01,  7.0743e-11],
         [-3.3576e-01, -7.5009e-11],
         [-2.7123e-01, -3.5793e-11],
         [ 1.7020e-01, -5.5290e-11],
         [ 9.5707e-02, -8.3681e-11],
         [-2.2416e-01, -4.0589e-11],
         [ 3.9455e-02, -6.7173e-11],
         [-3.2700e-01, -5.9575e-11],
         [-3.1624e-01,  2.5560e-10],
         [ 2.7330e-01, -2.0979e-10],
         [ 1.3217e-01, -5.6469e-11],
         [-1.9806e-01, -1.3773e-10],
         [-1.7177e-01,  1.1836e-10]], requires_grad=True))

분석 2(GCN): amt, time

# 'trans_date_trans_time' 열을 날짜/시간 형식으로 파싱
df50['trans_date_trans_time'] = pd.to_datetime(df50['trans_date_trans_time'])

# Unix 타임스탬프로 변환
df50['trans_date_trans_time'] = (df50['trans_date_trans_time'] - pd.Timestamp("1970-01-01")) // pd.Timedelta('1s')

# 나머지 코드는 이전과 동일하게 유지
x = torch.tensor(df50[['amt', 'trans_date_trans_time']].values, dtype=torch.float)
y = torch.tensor(df50['is_fraud'], dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index=edge_index_selected, y=y, train_mask=train_mask, test_mask=test_mask)
data


torch.manual_seed(202250926)
class GCN1(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(data.num_node_features, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN1()
optimizer = torch.optim.Adam(model.parameters(), lr=0.05, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

    model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results2= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석2'])
_results2

	accuracy_score	precision_score	recall_score	f1_score
분석2	0.897436	0.855798	0.958471	0.904229

---

분석 3(GCN): amt, category

np.count_nonzero(df50['category'].unique())

14

category_map = {category: index for index, category in enumerate(df50['category'].unique())}
df50['category'] = df50['category'].map(category_map)


x = torch.tensor(df50[['amt', 'category']].values, dtype=torch.float)
y = torch.tensor(df50['is_fraud'], dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index=edge_index_selected, y=y, train_mask=train_mask, test_mask=test_mask)
data


torch.manual_seed(202250926)
class GCN1(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(data.num_node_features, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN1()
optimizer = torch.optim.Adam(model.parameters(), lr=0.05, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

    model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results3= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석3'])
_results3

	accuracy_score	precision_score	recall_score	f1_score
분석3	0.899767	0.861905	0.954515	0.905849

---

분석 4(GCN): amt, distance

def haversine(lat1, lon1, lat2, lon2):
    # 지구의 반지름 (미터)
    radius = 6371.0

    # 라디안으로 변환
    lat1 = np.radians(lat1)
    lon1 = np.radians(lon1)
    lat2 = np.radians(lat2)
    lon2 = np.radians(lon2)

    # Haversine 공식 계산
    dlon = lon2 - lon1
    dlat = lat2 - lat1
    a = np.sin(dlat / 2)**2 + np.cos(lat1) * np.cos(lat2) * np.sin(dlon / 2)**2
    c = 2 * np.arctan2(np.sqrt(a), np.sqrt(1 - a))
    distance = radius * c

    return distance

# 데이터프레임(df50)에서 고객 위치 및 상점 위치의 위도와 경도 추출
customer_lat = df50['lat']
customer_lon = df50['long']
store_lat = df50['merch_lat']
store_lon = df50['merch_long']

# 거리 계산
distances = haversine(customer_lat, customer_lon, store_lat, store_lon)

# 거리를 데이터프레임에 추가
df50['distance_km'] = distances

x = torch.tensor(df50[['amt', 'distance_km']].values, dtype=torch.float)
y = torch.tensor(df50['is_fraud'], dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index=edge_index_selected, y=y, train_mask=train_mask, test_mask=test_mask)
data


torch.manual_seed(202250926)
class GCN1(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(data.num_node_features, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN1()
optimizer = torch.optim.Adam(model.parameters(), lr=0.05, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

    model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results4= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석4'])
_results4

	accuracy_score	precision_score	recall_score	f1_score
분석4	0.913753	0.866978	0.979565	0.919839

분석 5(GCN): amt, time, category

category_map = {category: index for index, category in enumerate(df50['category'].unique())}
df50['category'] = df50['category'].map(category_map)

df50['trans_date_trans_time'] = pd.to_datetime(df50['trans_date_trans_time'])
df50['trans_date_trans_time'] = (df50['trans_date_trans_time'] - pd.Timestamp("1970-01-01")) // pd.Timedelta('1s')

x = torch.tensor(df50[['amt', 'category', 'trans_date_trans_time']].values, dtype=torch.float)
y = torch.tensor(df50['is_fraud'], dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index=edge_index_selected, y=y, train_mask=train_mask, test_mask=test_mask)
data


torch.manual_seed(202250926)
class GCN1(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(data.num_node_features, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN1()
optimizer = torch.optim.Adam(model.parameters(), lr=0.05, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

    model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results5= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석5'])
_results5

	accuracy_score	precision_score	recall_score	f1_score
분석5	0.906094	0.866904	0.961767	0.911875

---

---

분석 6(GCN): amt, time, distance

df50['trans_date_trans_time'] = pd.to_datetime(df50['trans_date_trans_time'])
df50['trans_date_trans_time'] = (df50['trans_date_trans_time'] - pd.Timestamp("1970-01-01")) // pd.Timedelta('1s')

x = torch.tensor(df50[['amt', 'trans_date_trans_time', 'distance_km']].values, dtype=torch.float)
y = torch.tensor(df50['is_fraud'], dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index=edge_index_selected, y=y, train_mask=train_mask, test_mask=test_mask)
data


torch.manual_seed(202250926)
class GCN1(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(data.num_node_features, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN1()
optimizer = torch.optim.Adam(model.parameters(), lr=0.05, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

    model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results6= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석6'])
_results6

	accuracy_score	precision_score	recall_score	f1_score
분석6	0.913753	0.866978	0.979565	0.919839

---

분석 7(GCN): amt, category, distance

x = torch.tensor(df50[['amt', 'category', 'distance_km']].values, dtype=torch.float)
y = torch.tensor(df50['is_fraud'], dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index=edge_index_selected, y=y, train_mask=train_mask, test_mask=test_mask)
data


torch.manual_seed(202250926)
class GCN1(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(data.num_node_features, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN1()
optimizer = torch.optim.Adam(model.parameters(), lr=0.05, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

    model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results7= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석7'])
_results7

	accuracy_score	precision_score	recall_score	f1_score
분석7	0.914752	0.868498	0.979565	0.920694

---

분석 8(GCN): amt, time, category, distance

x = torch.tensor(df50[['amt', 'trans_date_trans_time', 'category', 'distance_km']].values, dtype=torch.float)
y = torch.tensor(df50['is_fraud'], dtype=torch.int64)
data = torch_geometric.data.Data(x=x, edge_index=edge_index_selected, y=y, train_mask=train_mask, test_mask=test_mask)
data


torch.manual_seed(202250926)
class GCN1(torch.nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = GCNConv(data.num_node_features, 32)
        self.conv2 = GCNConv(32,2)

    def forward(self, data):
        x, edge_index = data.x, data.edge_index

        x = self.conv1(x, edge_index)
        x = F.relu(x)
        x = F.dropout(x, training=self.training)
        x = self.conv2(x, edge_index)

        return F.log_softmax(x, dim=1)

X = (data.x[data.train_mask]).numpy()
XX = (data.x[data.test_mask]).numpy()
y = (data.y[data.train_mask]).numpy()
yy = (data.y[data.test_mask]).numpy()

model = GCN1()
optimizer = torch.optim.Adam(model.parameters(), lr=0.05, weight_decay=5e-4)
model.train()
for epoch in range(400):
    optimizer.zero_grad()
    out = model(data)
    loss = F.nll_loss(out[data.train_mask], data.y[data.train_mask])
    loss.backward()
    optimizer.step()

    model.eval()

pred = model(data).argmax(dim=1)
yyhat = pred[data.test_mask]

metrics = [sklearn.metrics.accuracy_score,
           sklearn.metrics.precision_score,
           sklearn.metrics.recall_score,
           sklearn.metrics.f1_score]

_results8= pd.DataFrame({m.__name__:[m(yy,yyhat).round(6)] for m in metrics},index=['분석8'])
_results8

	accuracy_score	precision_score	recall_score	f1_score
분석8	0.91342	0.861415	0.987475	0.920147