import warnings
import numpy as np
np.random.seed(42)
warnings.filterwarnings("ignore")
 
 

# Data setup (Restaurant Tips Dataset)
import seaborn as sns
df = sns.load_dataset('tips')
print(df.shape)
print(df.head(3))

# Vectorized math: add a "total" column
df['total'] = df['total_bill'] + df['tip']
print(df[['total_bill','tip','total']].head(3))

# Vectorized math with conditions
df['big_tip'] = df['tip'] > 5
print(df[['tip','big_tip']].head())

# Using lambda with apply() to tip column
df['tip_twice'] = df['tip'].apply(lambda x: x * 2)
print(df[['tip','tip_twice']].head())

# Conditional logic inside lambda
df['service_label'] = df['tip'].apply(lambda x: 'generous' if x > 5 else 'normal')
print(df[['tip','service_label']].head(6))

# Lambda with multiple columns via apply (axis=1)
df['bill_ratio'] = df.apply(lambda row: row['tip'] / row['total_bill'], axis=1)
print(df[['total_bill','tip','bill_ratio']].head(4))

# Vectorized string operation: add "$" to tip column
df['tip_str'] = df['tip'].astype(str) + " $"
print(df[['tip','tip_str']].head(5))

# Using .map() with a function
def bonus(size):
        if size >= 5:
                        return 1
        else:
                        return 0
df['party_bonus'] = df['size'].map(bonus)
print(df[['size','party_bonus']].head(6))

# Using np.where() for vectorized conditional assignment
import numpy as np
df['meal_value'] = np.where(df['total_bill'] > 30, 'expensive', 'standard')
print(df[['total_bill','meal_value']].head(7))

# Create tip_percent column (vectorized)
df['tip_percent'] = (df['tip'] / df['total_bill'] * 100).round(1)
print(df[['tip','total_bill','tip_percent']].head(8))

# Create tip_level column (lambda + apply)
df['tip_level'] = df['tip_percent'].apply(lambda x: 'high' if x >= 20 else 'normal')
print(df[['tip_percent','tip_level']].head(10))

# Vectorized math with missing values
df_missing = df.copy()
df_missing.loc[0,'tip'] = np.nan  # Set one tip to NaN
df_missing['tip_plus_five'] = df_missing['tip'] + 5
print(df_missing[['tip','tip_plus_five']].head(3))

# Clean missing values with .fillna() and re-apply math
df_missing['tip_filled'] = df_missing['tip'].fillna(0)
df_missing['new_total'] = df_missing['tip_filled'] + df_missing['total_bill']
print(df_missing[['tip','tip_filled','new_total']].head(4))

# Challenge practice: flag rows where tip is above mean
mean_tip = df['tip'].mean()
df['above_average_tip'] = df['tip'] > mean_tip
print(f"Average tip: {mean_tip:.2f}")
print(df[['tip','above_average_tip']].sample(5, random_state=42))