import matplotlib.pyplot as plt
from sklearn.manifold import TSNE
from imblearn.over_sampling import SMOTE

# Set the font family for matplotlib
plt.rcParams['font.family'] = 'Verdana'

# Split your data into features and labels
X = df.drop('target', axis=1)
y = df['target']

# Define the SMOTE resampling method
smote = SMOTE(random_state=42)

# Resample the data using SMOTE
X_resampled, y_resampled = smote.fit_resample(X, y)

# Use TSNE for dimensionalilty reduction
X_tsne = TSNE(n_components=2).fit_transform(X)
X_resampled_tsne = TSNE(n_components=2).fit_transform(X_resampled)

# Get the class ratios
original_class_ratio = np.bincount(y) / len(y)
resampled_class_ratio = np.bincount(y_resampled) / len(y_resampled)

# Create subplots
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 6))

# Plot the original data
ax1.scatter(X_tsne[:, 0], X_tsne[:, 1], c=y, cmap='rainbow',alpha=0.4)
ax1.set_title(f'Original DatanClass Ratio: {original_class_ratio[1] * 100:.1f} : {original_class_ratio[0] * 100:.1f}')

# Plot the resampled data
ax2.scatter(X_resampled_tsne[:, 0], X_resampled_tsne[:, 1], c=y_resampled, cmap='rainbow',alpha=0.4)
ax2.set_title(f'Resampled DatanClass Ratio: {resampled_class_ratio[1] * 100:.0f} : {resampled_class_ratio[0] * 100:.0f}')

plt.show()