train_svm.py

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler, LabelEncoder
from sklearn.svm import SVC
from sklearn.metrics import confusion_matrix, classification_report, accuracy_score
import matplotlib.pyplot as plt
import seaborn as sns


# 1. Load the dataset

print("🔹 Loading extracted features...")
data = pd.read_csv("features.csv")

# Separate features (X) and labels (y)
X = data.drop(columns=['label'])
y = data['label']

# Encode string labels into numbers
le = LabelEncoder()
y = le.fit_transform(y)


# 2. Train-test split

X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42, stratify=y
)


# 3. Normalize features

scaler = StandardScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)


# 4. Train SVM model

print("Training SVM classifier...")
model = SVC(kernel='rbf', C=10, gamma='scale')
model.fit(X_train, y_train)

# 5. Evaluate model

print("\n Model training complete!")

y_pred = model.predict(X_test)

acc = accuracy_score(y_test, y_pred)
print(f"\n Accuracy: {acc * 100:.2f}%\n")

print(" Classification Report:")
print(classification_report(y_test, y_pred, target_names=le.classes_))


# 6. Confusion Matrix (Beautiful Visualization)

cm = confusion_matrix(y_test, y_pred)
genres = le.classes_

plt.figure(figsize=(10, 8))
sns.set(font_scale=1.1)
sns.heatmap(
    cm, annot=True, fmt="d", cmap="crest", linewidths=0.8, 
    xticklabels=genres, yticklabels=genres, cbar_kws={"shrink": 0.8}
)

plt.title("🎵 Confusion Matrix - Music Genre Classification (SVM)", fontsize=16, weight='bold', pad=15)
plt.xlabel("Predicted Labels", fontsize=13)
plt.ylabel("Actual Labels", fontsize=13)

# Rotate x labels for clarity
plt.xticks(rotation=30, ha='right')
plt.yticks(rotation=0)

# Add overall accuracy text
plt.text(len(genres)-2, -1.2, f"Accuracy: {acc*100:.2f}%", fontsize=12, color="green", weight='bold')

plt.tight_layout()
plt.show()