Statistical visualizations for distribution and drift analysis

                  import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

train_df = pd.read_parquet('train_features.parquet')
prod_df = pd.read_parquet('production_features.parquet')

train_df['dataset'] = 'train'
prod_df['dataset'] = 'production'

combined = pd.concat([
    train_df[['transaction_amount', 'dataset']],
    prod_df[['transaction_amount', 'dataset']],
])

sns.kdeplot(data=combined, x='transaction_amount', hue='dataset', fill=True, common_norm=False)
plt.xscale('log')
plt.title('Distribution drift for transaction_amount')
plt.tight_layout()
plt.show()