import warnings
warnings.filterwarnings("ignore")

# We will use the Tips dataset for demonstration
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
np.random.seed(42)

df = sns.load_dataset('tips')
print(df.shape)
print(df.head(3))

# Plotting the total bill amounts as a histogram
df['total_bill'].plot(kind='hist', bins=20, edgecolor='black', figsize=(8,4))
plt.title('Distribution of Total Bill Amounts')
plt.xlabel('Total Bill')
plt.ylabel('Frequency')
plt.show()

# Seaborn: a prettier histogram with a kde curve
sns.histplot(df['total_bill'], bins=20, kde=True, color='skyblue', edgecolor='black')
plt.title('Total Bill Distribution (Seaborn)')
plt.xlabel('Total Bill')
plt.ylabel('Count')
plt.show()

# Scatter plot: Tip vs. Total Bill
plt.figure(figsize=(8,5))
plt.scatter(df['total_bill'], df['tip'], alpha=0.7)
plt.title('Scatter Plot of Total Bill vs Tip Amount')
plt.xlabel('Total Bill')
plt.ylabel('Tip')
plt.show()

# Seaborn scatter plot with automatic regression line
sns.lmplot(x='total_bill', y='tip', data=df, height=5, aspect=1.5)
plt.title('Seaborn: Tips vs. Total Bill with Trend Line')
plt.xlabel('Total Bill')
plt.ylabel('Tip')
plt.show()

# Plotting by category: Average tip by day
avg_tip_per_day = df.groupby('day')['tip'].mean()
avg_tip_per_day.plot(kind='bar', color='orange', figsize=(6,4))
plt.title('Average Tip by Day of Week')
plt.xlabel('Day of Week')
plt.ylabel('Average Tip')
plt.show()

# Seaborn barplot: Average tip by sex and smoker status
plt.figure(figsize=(6,4))
sns.barplot(x='sex', y='tip', hue='smoker', data=df, ci=None, palette='Set2')
plt.title('Average Tip by Gender and Smoking Status')
plt.xlabel('Gender')
plt.ylabel('Average Tip')
plt.show()

# Heatmap: Correlation between numeric columns
corr = df.corr(numeric_only=True)
sns.heatmap(corr, annot=True, cmap='coolwarm', fmt='.2f')
plt.title('Heatmap of Feature Correlations')
plt.show()

# Pairplot: Visualize all feature relationships at once
sns.pairplot(df, diag_kind='kde', hue='sex', palette='muted', plot_kws={'alpha':0.7})
plt.suptitle('Pairplot of Tips Dataset (Colored by Gender)', y=1.02)
plt.show()

# Changing style and adding annotations
sns.set_style('whitegrid')
plt.figure(figsize=(7,5))
sns.boxplot(x='day', y='total_bill', data=df, palette='pastel')
plt.title('Total Bill by Day of Week')
plt.xlabel('Day of Week')
plt.ylabel('Total Bill')
plt.annotate('Outlier?', xy=(1, 50), xytext=(1.6, 45), arrowprops=dict(arrowstyle='->'))
plt.show()

# Saving your plots to disk
fig, ax = plt.subplots(figsize=(6,4))
sns.histplot(df['tip'], bins=15, ax=ax, color='purple')
ax.set_title('Histogram of Tips')
ax.set_xlabel('Tip Amount')
ax.set_ylabel('Count')
fig.tight_layout()
fig.savefig('tips_histogram.png')

# EXERCISE: Customize a plot yourself
plt.figure(figsize=(8,4))
sns.violinplot(x='sex', y='tip', data=df, palette='husl')
plt.title('Tip Distribution by Gender')
plt.xlabel('Gender')
plt.ylabel('Tip Amount')
plt.show()

# Create a column labeling each entry as 'Weekday' or 'Weekend'
df['day_type'] = df['day'].apply(lambda x: 'Weekend' if x in ['Sat', 'Sun'] else 'Weekday')
# Group by day_type and calculate mean tip
weekday_vs_weekend = df.groupby('day_type')['tip'].mean().reset_index()
sns.barplot(x='day_type', y='tip', data=weekday_vs_weekend, palette='viridis')
plt.title('Average Tip: Weekday vs Weekend')
plt.xlabel('Day Type')
plt.ylabel('Average Tip')
plt.show()

# MINI-PROJECT PART 2: Visualizing tip percentage by party size
df['tip_pct'] = 100 * df['tip'] / df['total_bill']
sns.boxplot(x='size', y='tip_pct', data=df, palette='rocket')
plt.title('Tip Percentage by Party Size')
plt.xlabel('Party Size')
plt.ylabel('Tip % of Total Bill')
plt.ylim(0,50)
plt.show()

# Best practice: Use style context manager for publication-ready charts
with sns.axes_style('darkgrid'):
    plt.figure(figsize=(7,4))
    sns.countplot(x='day', hue='smoker', data=df, palette='deep')
    plt.title('Smoker Count by Day of Week')
    plt.xlabel('Day of Week')
    plt.ylabel('Number of Smokers')
    plt.show()
    

# Troubleshooting: What if you get a confusing plot or an error?
# Example: Misspelling a column name
try:
    sns.boxplot(x='day', y='full_bill', data=df)
except Exception as e:
    print('Error:', e)
    

# CHALLENGE: Plot tip percentage by time and smoker status
plt.figure(figsize=(8,5))
sns.violinplot(x='time', y='tip_pct', hue='smoker', split=True, data=df, palette='Set3')
plt.title('Tip Percentage by Meal Time and Smoker')
plt.xlabel('Meal Time')
plt.ylabel('Tip % of Total Bill')
plt.ylim(0,50)
plt.show()