[Kaggle] Heart Disease Cleveland

data

• Age: Patients Age in years (Numeric)
• Sex: Gender (Male : 1; Female : 0) (Nominal)
• cp: Type of chest pain experienced by patient. This term categorized into 4 category.
• 0 typical angina, 1 atypical angina, 2 non- anginal pain, 3 asymptomatic (Nominal)
• trestbps: patient’s level of blood pressure at resting mode in mm/HG (Numerical)
• chol: Serum cholesterol in mg/dl (Numeric)
• fbs: Blood sugar levels on fasting > 120 mg/dl represents as 1 in case of true and 0 as false (Nominal)
• restecg: Result of electrocardiogram while at rest are represented in 3 distinct values
• 0 : Normal 1: having ST-T wave abnormality (T wave inversions and/or ST elevation or depression of > 0.05 mV)
• 2: showing probable or definite left ventricular hypertrophyby Estes’ criteria (Nominal)
• thalach: Maximum heart rate achieved (Numeric)
• exang: Angina induced by exercise 0 depicting NO 1 depicting Yes (Nominal)
• oldpeak: Exercise induced ST-depression in relative with the state of rest (Numeric)
• slope: ST segment measured in terms of slope during peak exercise
• 0: up sloping; 1: flat; 2: down sloping(Nominal)
• ca: The number of major vessels (0–3)(nominal)
• thal: A blood disorder called thalassemia
• 0: NULL 1: normal blood flow 2: fixed defect (no blood flow in some part of the heart) 3: reversible defect (a blood flow is observed but it is not normal(nominal)
• target: It is the target variable which we have to predict 1 means patient is suffering from heart disease and 0 means patient is normal.

import pandas as pd
import numpy as np

import torch
from sklearn.preprocessing import StandardScaler
from sklearn.neighbors import NearestNeighbors

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

from sklearn.cluster import KMeans

df = pd.read_csv('../../../delete/Heart_disease_cleveland_new.csv')

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 303 entries, 0 to 302
Data columns (total 14 columns):
 #   Column    Non-Null Count  Dtype  
---  ------    --------------  -----  
 0   age       303 non-null    int64  
 1   sex       303 non-null    int64  
 2   cp        303 non-null    int64  
 3   trestbps  303 non-null    int64  
 4   chol      303 non-null    int64  
 5   fbs       303 non-null    int64  
 6   restecg   303 non-null    int64  
 7   thalach   303 non-null    int64  
 8   exang     303 non-null    int64  
 9   oldpeak   303 non-null    float64
 10  slope     303 non-null    int64  
 11  ca        303 non-null    int64  
 12  thal      303 non-null    int64  
 13  target    303 non-null    int64  
dtypes: float64(1), int64(13)
memory usage: 33.3 KB

df.head(5)

	age	sex	cp	trestbps	chol	fbs	restecg	thalach	exang	oldpeak	slope	ca	thal	target
0	63	1	0	145	233	1	2	150	0	2.3	2	0	2	0
1	67	1	3	160	286	0	2	108	1	1.5	1	3	1	1
2	67	1	3	120	229	0	2	129	1	2.6	1	2	3	1
3	37	1	2	130	250	0	0	187	0	3.5	2	0	1	0
4	41	0	1	130	204	0	2	172	0	1.4	0	0	1	0

X = df.drop(columns=['target']).values

X = StandardScaler().fit_transform(X)

knn = NearestNeighbors(n_neighbors=5)

knn

NearestNeighbors()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

knn.fit(X)

NearestNeighbors()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

edges = knn.kneighbors(X,return_distance=False)

edges

array([[  0, 124,  49, 196, 139],
       [  1,  37, 155, 235,  65],
       [  2,  24, 206,  76,  62],
       ...,
       [300, 250,  64, 177, 127],
       [301, 135, 125, 230, 276],
       [302, 269, 283,  35, 190]])

edge_index = []
for i in range(len(edges)):
    for j in edges[i]:
        edge_index.append([i,j])

len(edge_index)

1515

torch.tensor(edge_index).shape

torch.Size([1515, 2])

torch.tensor(edge_index).t().shape

torch.Size([2, 1515])

torch.tensor(edge_index).t().contiguous().shape

torch.Size([2, 1515])

edge_index = torch.tensor(edge_index).t().contiguous()

x = torch.tensor(X, dtype=torch.float)

x

tensor([[ 0.9487,  0.6862, -2.2518,  ...,  2.2746, -0.7111,  0.1762],
        [ 1.3920,  0.6862,  0.8780,  ...,  0.6491,  2.5049, -0.8708],
        [ 1.3920,  0.6862,  0.8780,  ...,  0.6491,  1.4329,  1.2232],
        ...,
        [ 0.2838,  0.6862,  0.8780,  ...,  0.6491,  0.3609,  1.2232],
        [ 0.2838, -1.4573, -1.2085,  ...,  0.6491,  0.3609, -0.8708],
        [-1.8217,  0.6862, -0.1653,  ..., -0.9764, -0.7111, -0.8708]])

class GCNencode(torch.nn.Module):
    def __init__(self, in_channels):
        super().__init__()
        self.conv1 = GCNConv(in_channels, 64)
        self.conv2 = GCNConv(64,32)
        self.conv3 = GCNConv(32,13)
    
    def forward(self, x, edge_index):
        x = self.conv1(x,edge_index)
        x = F.relu(x)
        x = self.conv2(x,edge_index)
        x = F.relu(x)
        x = self.conv3(x,edge_index)
        return x   

x.shape

torch.Size([303, 13])

x.shape[0]

303

x.shape[1]

13

model = GCNencode(x.shape[1])

model

GCNencode(
  (conv1): GCNConv(13, 64)
  (conv2): GCNConv(64, 32)
  (conv3): GCNConv(32, 13)
)

optimizer = torch.optim.Adam(model.parameters(), lr = 0.01)

x[:,:13]

tensor([[ 0.9487,  0.6862, -2.2518,  ...,  2.2746, -0.7111,  0.1762],
        [ 1.3920,  0.6862,  0.8780,  ...,  0.6491,  2.5049, -0.8708],
        [ 1.3920,  0.6862,  0.8780,  ...,  0.6491,  1.4329,  1.2232],
        ...,
        [ 0.2838,  0.6862,  0.8780,  ...,  0.6491,  0.3609,  1.2232],
        [ 0.2838, -1.4573, -1.2085,  ...,  0.6491,  0.3609, -0.8708],
        [-1.8217,  0.6862, -0.1653,  ..., -0.9764, -0.7111, -0.8708]])

for epoch in range(100):
    model.train()
    optimizer.zero_grad()
    
    z = model(x, edge_index)
    loss = F.mse_loss(z,x[:,:13])
    loss.backward()
    optimizer.step()

model.eval()

GCNencode(
  (conv1): GCNConv(13, 64)
  (conv2): GCNConv(64, 32)
  (conv3): GCNConv(32, 13)
)

embeddings = model(x,edge_index).detach().numpy()

embeddings

array([[ 0.93777424,  1.1018754 , -2.2913668 , ...,  1.904715  ,
        -0.5608    , -0.23465104],
       [ 1.4721875 ,  0.82115763, -0.11787386, ...,  0.45280465,
         2.554476  , -0.60368335],
       [ 0.7943949 ,  0.8147143 ,  0.9416017 , ...,  0.30953175,
         1.5317379 ,  1.3029628 ],
       ...,
       [ 0.28482667,  0.437107  ,  0.8282606 , ...,  1.0481589 ,
        -0.33025122,  0.8082806 ],
       [ 0.0184263 , -1.4629475 , -1.0692531 , ...,  0.26454   ,
        -0.32980496, -0.8803086 ],
       [-1.9536707 ,  0.6859256 , -0.09410449, ..., -1.2229915 ,
        -0.93635803, -0.6885667 ]], dtype=float32)

embeddings.shape

torch.Size([303, 13])

kmeans = KMeans(n_clusters=2)

kmeans

KMeans(n_clusters=2)

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.