CWHFDL

A package for data classification with WHFDL.

This package is part of "WHFDL: an explainable method based on World Hyper-heuristic and Fuzzy Deep Learning approaches for gastric cancer detection using metabolomics data" article's experiment.

Example:

import torch
import torch.nn as nn
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from torch.utils.data import DataLoader, TensorDataset, random_split
from sklearn.preprocessing import StandardScaler
from sklearn.metrics import roc_curve, auc, confusion_matrix, classification_report

import CWHFDL as cw

cw.set_seed(42)

df = pd.read_csv('CGMAIN1.csv')
X = df.drop(columns='state').values
y = df['state'].values

scaler = StandardScaler()
X = scaler.fit_transform(X)

X_tensor = torch.tensor(X, dtype=torch.float32)
y_tensor = torch.tensor(y, dtype=torch.long)

dataset = TensorDataset(X_tensor, y_tensor)
train_size = int(0.8 * len(dataset))
val_size = len(dataset) - train_size
train_dataset, val_dataset = random_split(dataset, [train_size, val_size])

train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True)
test_loader = DataLoader(val_dataset, batch_size=32, shuffle=False)

model = cw.FDNN(
    in_features=X.shape[1],
    hidden_dim=128,
    num_memberships=3,
    num_classes=len(np.unique(y)),
    dropout_rate=0.1
)

cw.initialize_fuzzy_layer(model.fuzzy, train_loader)

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

model.train()
for epoch in range(100):
    total_loss = 0
    for x_batch, y_batch in train_loader:
        optimizer.zero_grad()
        outputs = model(x_batch)
        loss = criterion(outputs, y_batch)
        loss.backward()
        optimizer.step()
        total_loss += loss.item()
    print(f"Epoch {epoch + 1}: Loss = {total_loss:.4f}")

model.eval()
all_preds, all_probs, all_labels = [], [], []
with torch.no_grad():
    for x_batch, y_batch in test_loader:
        outputs = model(x_batch)
        probs = torch.softmax(outputs, dim=1)
        preds = torch.argmax(probs, dim=1)
        all_preds.extend(preds.numpy())
        all_probs.extend(probs[:, 1].numpy())
        all_labels.extend(y_batch.numpy())

print(
    classification_report(
        all_labels,
        all_preds,
        target_names=[
            'Negative',
            'Positive']))

conf_matrix = confusion_matrix(all_labels, all_preds)
plt.figure(figsize=(8, 6))
sns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues',
            xticklabels=['Predicted Negative', 'Predicted Positive'],
            yticklabels=['Actual Negative', 'Actual Positive'])
plt.title('WHFDL  Confusion Matrix', fontsize=16, fontweight='bold')
plt.xlabel('Predicted Label', fontsize=14)
plt.ylabel('True Label', fontsize=14)
plt.show()

fpr, tpr, thresholds = roc_curve(all_labels, all_probs)
roc_auc = auc(fpr, tpr)

plt.figure(figsize=(8, 6))
plt.plot(fpr, tpr, color='blue', lw=2, label=f'ROC Curve (AUC = {roc_auc:.4f}')
plt.plot([0, 1], [0, 1], color='red', linestyle='--', label='Random Guessing')
plt.xlabel('False Positive Rate', fontsize=14)
plt.ylabel('True Positive Rate', fontsize=14)
plt.title('WHFDL (ROC)', fontsize=16, fontweight='bold')
plt.legend(loc="lower right")
plt.grid(True)
plt.show()

You can find the "CGMAIN1.csv" here.