🎨 Seaborn

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

# Set global theme — do this once at the top
sns.set_theme(style='whitegrid', palette='muted', font_scale=1.1)

# Seaborn ships with practice datasets
tips  = sns.load_dataset('tips')
iris  = sns.load_dataset('iris')
titanic = sns.load_dataset('titanic')

print(tips.head())
print(f"\ntips shape: {tips.shape}")
print(tips.dtypes)

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

styles   = ['darkgrid','whitegrid','dark','white','ticks']
palettes = ['deep','muted','pastel','bright','dark','colorblind']

fig, axes = plt.subplots(2, 3, figsize=(12, 6))
for ax, style in zip(axes[0], styles[:3]):
    sns.barplot(x=['A','B','C'], y=[3,5,4], ax=ax, palette='deep')
    ax.set_title(f'style={style}')

for ax, pal in zip(axes[1], palettes[:3]):
    sns.barplot(x=['A','B','C'], y=[3,5,4], ax=ax, palette=pal)
    ax.set_title(f'palette={pal}')

plt.tight_layout()
plt.savefig('styles_palettes.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np

# Create and inspect custom palettes
blues   = sns.color_palette("Blues", 6)
custom  = sns.color_palette(["#e74c3c","#3498db","#2ecc71","#f39c12"])
husl    = sns.color_palette("husl", 8)

fig, axes = plt.subplots(1, 3, figsize=(12, 2))
sns.palplot(blues,  ax=axes[0] if hasattr(sns,'palplot') else None)
for ax, (pal, name) in zip(axes, [(blues,'Blues-6'),(custom,'Custom-4'),(husl,'HUSL-8')]):
    for j, c in enumerate(pal):
        ax.add_patch(plt.Rectangle((j, 0), 1, 1, color=c))
    ax.set_xlim(0, len(pal)); ax.set_ylim(0, 1)
    ax.set_xticks([]); ax.set_yticks([])
    ax.set_title(name)

plt.suptitle('Color Palette Examples', y=1.05)
plt.tight_layout()
plt.savefig('palettes_custom.png', dpi=80)
plt.close()
print("Blues palette (RGB):", [tuple(round(v,2) for v in c) for c in blues])

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

# --- axes-level function: returns an Axes, fits into any subplot grid ---
fig, axes = plt.subplots(1, 2, figsize=(12, 4))

sns.boxplot(data=tips, x='day', y='total_bill', palette='pastel', ax=axes[0])
axes[0].set_title('Axes-level: sns.boxplot (axes[0])')
# sns.despine removes top/right spines for a cleaner look
sns.despine(ax=axes[0], offset=8, trim=True)

# A second axes-level call on the other subplot
sns.stripplot(data=tips, x='day', y='total_bill',
              color='steelblue', alpha=0.4, jitter=True, ax=axes[1])
axes[1].set_title('Axes-level: sns.stripplot (axes[1])')
sns.despine(ax=axes[1])

plt.suptitle('Axes-level functions — full subplot control', fontsize=13, y=1.02)
plt.tight_layout()
plt.savefig('setup_despine_axes_level.png', dpi=80)
plt.close()

# --- figure-level function: creates its OWN figure / FacetGrid ---
# You cannot pass ax= to figure-level functions (displot, catplot, relplot, etc.)
g = sns.displot(data=tips, x='total_bill', col='time',
                kind='hist', kde=True, bins=20,
                height=3.5, aspect=1.1, palette='Set2')
g.set_titles('{col_name}')
g.figure.suptitle('Figure-level: sns.displot (owns its own Figure)', y=1.04)
plt.tight_layout()
plt.savefig('setup_despine_figure_level.png', dpi=80)
plt.close()
print("despine + figure-level vs axes-level demo saved.")

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(11, 4))

# Histogram with KDE overlay
sns.histplot(data=tips, x='total_bill', bins=25,
             kde=True, ax=axes[0], color='steelblue')
axes[0].set_title('Total Bill Distribution')

# KDE only — compare two groups
sns.kdeplot(data=tips, x='tip', hue='sex',
            fill=True, alpha=0.4, ax=axes[1])
axes[1].set_title('Tip Distribution by Gender')

plt.tight_layout()
plt.savefig('dist_hist_kde.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

# Separate histogram per day, colored by sex
g = sns.displot(
    data=tips, x='total_bill', hue='sex',
    col='day', kind='hist', kde=True,
    bins=15, height=3.5, aspect=0.9,
    palette='Set2'
)
g.set_titles('{col_name}')
g.set_xlabels('Total Bill ($)')
plt.tight_layout()
plt.savefig('dist_displot.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(11, 4))

# ECDF — empirical cumulative distribution
sns.ecdfplot(data=tips, x='total_bill', hue='time',
             palette='Set1', linewidth=2, ax=axes[0])
axes[0].set_title('ECDF of Total Bill by Meal Time')
axes[0].set_xlabel('Total Bill ($)')

# Rug plot layered under KDE
sns.kdeplot(data=tips, x='total_bill', hue='smoker',
            fill=True, alpha=0.3, ax=axes[1])
sns.rugplot(data=tips, x='total_bill', hue='smoker',
            height=0.06, ax=axes[1])
axes[1].set_title('KDE + Rug: Total Bill by Smoker')

plt.tight_layout()
plt.savefig('dist_ecdf_rug.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
sns.set_theme(style='whitegrid')

# Synthetic spend data for three customer tiers
tiers = {'Bronze': np.random.normal(30, 10, 300),
         'Silver': np.random.normal(70, 15, 200),
         'Gold':   np.random.normal(130, 25, 100)}
rows = []
for tier, vals in tiers.items():
    for v in vals.clip(0):
        rows.append({'tier': tier, 'spend': round(v, 2)})
df = pd.DataFrame(rows)

fig, axes = plt.subplots(1, 2, figsize=(13, 4))

# Left: overlaid histograms normalised to density so groups are comparable
for tier, color in [('Bronze','#cd7f32'),('Silver','#aaa9ad'),('Gold','#ffd700')]:
    subset = df[df['tier'] == tier]
    sns.histplot(data=subset, x='spend', stat='density',
                 bins=25, alpha=0.45, color=color, label=tier,
                 kde=True, ax=axes[0])
axes[0].set_title('Spend Distribution by Tier (stat=density)')
axes[0].set_xlabel('Monthly Spend ($)')
axes[0].legend(title='Tier')

# Right: displot with kind='ecdf' — figure-level, drawn into a fresh figure
plt.tight_layout()
plt.savefig('dist_density_overlay.png', dpi=80)
plt.close()

g = sns.displot(data=df, x='spend', hue='tier',
                kind='ecdf', linewidth=2.5,
                palette={'Bronze':'#cd7f32','Silver':'#aaa9ad','Gold':'#ffd700'},
                height=4, aspect=1.6)
g.set_axis_labels('Monthly Spend ($)', 'Proportion')
g.figure.suptitle('ECDF of Spend by Customer Tier (displot kind=ecdf)', y=1.03)
plt.tight_layout()
plt.savefig('dist_ecdf_displot.png', dpi=80)
plt.close()
print("Density overlay and ECDF displot saved.")

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(11, 4))

# Mean tip by day + confidence interval
sns.barplot(data=tips, x='day', y='tip',
            hue='sex', palette='Set2', ax=axes[0])
axes[0].set_title('Avg Tip by Day & Gender')
axes[0].set_ylabel('Mean Tip ($)')

# Count of meals per day
sns.countplot(data=tips, x='day', hue='time',
              palette='pastel', ax=axes[1])
axes[1].set_title('Meal Count by Day & Time')

plt.tight_layout()
plt.savefig('cat_bar_count.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

# Compute mean tip per day — sort for readability
order = tips.groupby('day')['tip'].mean().sort_values(ascending=False).index

fig, ax = plt.subplots(figsize=(7, 4))
sns.barplot(data=tips, y='day', x='tip',
            order=order, palette='Blues_d', orient='h', ax=ax)
ax.set_xlabel('Average Tip ($)')
ax.set_title('Average Tip by Day of Week (sorted)')
plt.tight_layout()
plt.savefig('cat_hbar.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(7)
sns.set_theme(style='whitegrid')

# Simulated quarterly sales by product
products = ['Alpha', 'Beta', 'Gamma']
quarters = ['Q1', 'Q2', 'Q3', 'Q4']
rows = []
for q in quarters:
    for p in products:
        base = {'Alpha': 120, 'Beta': 85, 'Gamma': 160}[p]
        rows.append({'quarter': q, 'product': p,
                     'sales': np.random.normal(base, 15)})
df = pd.DataFrame(rows)

fig, ax = plt.subplots(figsize=(9, 4))
sns.barplot(data=df, x='quarter', y='sales', hue='product',
            palette='Set2', capsize=0.08, ax=ax)

# Value labels on bars
for bar in ax.patches:
    if bar.get_height() > 0:
        ax.text(bar.get_x() + bar.get_width()/2,
                bar.get_height() + 1,
                f'{bar.get_height():.0f}',
                ha='center', va='bottom', fontsize=7)

ax.set_title('Quarterly Sales by Product (mean ± 95% CI)')
ax.set_ylabel('Sales ($K)')
plt.tight_layout()
plt.savefig('cat_grouped_bar.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(21)
sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

# --- catplot: figure-level wrapper for all categorical plot kinds ---
# kind='bar' with estimator=median and dodge=True (default) for grouped bars
g = sns.catplot(
    data=tips, x='day', y='total_bill', hue='time',
    kind='bar',
    estimator=np.median,       # use median instead of default mean
    errorbar=('ci', 95),       # 95 % bootstrap CI
    dodge=True,                # bars side-by-side (not stacked)
    palette='Set1',
    capsize=0.10,
    height=4, aspect=1.4,
    order=['Thur', 'Fri', 'Sat', 'Sun']
)
g.set_axis_labels('Day', 'Median Total Bill ($)')
g.set_titles('Tip vs Bill — median estimator')
g.figure.suptitle('catplot: Median Total Bill by Day & Meal Time (dodge=True)', y=1.03)
plt.tight_layout()
plt.savefig('cat_catplot_median_dodge.png', dpi=80)
plt.close()

# --- second figure: countplot with dodge to compare two binary variables ---
fig, ax = plt.subplots(figsize=(8, 4))
sns.countplot(data=tips, x='day', hue='smoker',
              dodge=True, palette='Dark2',
              order=['Thur', 'Fri', 'Sat', 'Sun'], ax=ax)
ax.set_title('Diner Count by Day & Smoker Status (dodge=True)')
ax.set_ylabel('Count')
plt.tight_layout()
plt.savefig('cat_count_dodge.png', dpi=80)
plt.close()
print("catplot (median, dodge) and countplot saved.")

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(12, 4))

sns.boxplot(data=tips, x='day', y='total_bill',
            hue='time', palette='Set3', ax=axes[0])
axes[0].set_title('Total Bill — Box Plot')

sns.violinplot(data=tips, x='day', y='total_bill',
               hue='time', split=True,
               palette='Set2', inner='quartile', ax=axes[1])
axes[1].set_title('Total Bill — Violin Plot')

plt.tight_layout()
plt.savefig('cat_box_violin.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, ax = plt.subplots(figsize=(8, 5))

# Letter-value plot (boxenplot) for larger datasets
sns.boxenplot(data=tips, x='day', y='total_bill',
              palette='muted', ax=ax)

# Overlay raw points
sns.stripplot(data=tips, x='day', y='total_bill',
              color='black', size=3, alpha=0.3, ax=ax, jitter=True)

ax.set_title('Total Bill Distribution per Day (boxen + strip)')
ax.set_ylabel('Total Bill ($)')
plt.tight_layout()
plt.savefig('cat_boxen_strip.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(42)
sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

fig, axes = plt.subplots(1, 2, figsize=(13, 5))

# Left: violin per species for petal_length
sns.violinplot(data=iris, x='species', y='petal_length',
               palette='Set2', inner='box', ax=axes[0])
axes[0].set_title('Petal Length by Species (inner=box)')
axes[0].set_ylabel('Petal Length (cm)')

# Right: layered — violin + box + strip
sns.violinplot(data=iris, x='species', y='sepal_width',
               palette='pastel', inner=None, ax=axes[1])
sns.boxplot(data=iris, x='species', y='sepal_width',
            width=0.12, fliersize=0,
            boxprops=dict(facecolor='white', zorder=2), ax=axes[1])
sns.stripplot(data=iris, x='species', y='sepal_width',
              color='black', size=2.5, alpha=0.4, jitter=True, ax=axes[1])
axes[1].set_title('Sepal Width: Violin + Box + Strip')
axes[1].set_ylabel('Sepal Width (cm)')

plt.tight_layout()
plt.savefig('cat_violin_layered.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np

np.random.seed(7)
sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(13, 5))

# Left: swarmplot — all points plotted without overlap (small datasets)
sns.swarmplot(data=tips, x='day', y='total_bill',
              hue='sex', dodge=True,
              palette='Set2', size=4, ax=axes[0])
axes[0].set_title('swarmplot: Total Bill by Day & Sex')
axes[0].set_ylabel('Total Bill ($)')
axes[0].legend(title='Sex', loc='upper left')

# Right: pointplot — shows mean + CI as a connected dot plot
#   great for showing trends across ordered categories
sns.pointplot(data=tips, x='day', y='tip', hue='sex',
              dodge=0.3, linestyles=['--', '-'],
              markers=['o', 's'], palette='Set1',
              capsize=0.12, errorbar=('ci', 95),
              order=['Thur', 'Fri', 'Sat', 'Sun'],
              ax=axes[1])
axes[1].set_title('pointplot: Mean Tip by Day & Sex (95% CI)')
axes[1].set_ylabel('Mean Tip ($)')
axes[1].legend(title='Sex')

plt.tight_layout()
plt.savefig('cat_swarm_point.png', dpi=80)
plt.close()
print("swarmplot and pointplot saved.")

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, ax = plt.subplots(figsize=(8, 5))
sns.scatterplot(
    data=tips, x='total_bill', y='tip',
    hue='time', size='size',
    sizes=(40, 200), alpha=0.7,
    palette='Set1', ax=ax
)
ax.set_title('Tip vs Total Bill (size = party size)')
ax.set_xlabel('Total Bill ($)'); ax.set_ylabel('Tip ($)')
plt.tight_layout()
plt.savefig('scatter_hue_size.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, ax = plt.subplots(figsize=(7, 4))

# regplot — single regression line
sns.regplot(data=tips, x='total_bill', y='tip',
            scatter_kws=dict(alpha=0.4, s=30),
            line_kws=dict(color='red', linewidth=2), ax=ax)
ax.set_title('regplot: Tip vs Bill')
plt.tight_layout()
plt.savefig('scatter_regplot.png', dpi=80)
plt.close()

# lmplot — regression per group (returns FacetGrid)
g = sns.lmplot(data=tips, x='total_bill', y='tip',
               hue='smoker', palette='Set1',
               scatter_kws=dict(alpha=0.4, s=25),
               height=4, aspect=1.4)
g.set_axis_labels('Total Bill ($)', 'Tip ($)')
g.figure.suptitle('lmplot: Tip vs Bill by Smoker', y=1.02)
plt.tight_layout()
plt.savefig('scatter_lmplot.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(10)
sns.set_theme(style='whitegrid')

# Synthetic dataset with 3 categorical dimensions
n = 120
df = pd.DataFrame({
    'spend':    np.random.uniform(10, 500, n),
    'revenue':  np.random.uniform(50, 2000, n),
    'channel':  np.random.choice(['Search', 'Social', 'Email'], n),
    'region':   np.random.choice(['North', 'South', 'East'], n),
    'budget':   np.random.uniform(5, 50, n),
})

fig, ax = plt.subplots(figsize=(9, 5))
sns.scatterplot(
    data=df, x='spend', y='revenue',
    hue='channel',    # color
    size='budget',    # marker area
    style='region',   # marker shape
    sizes=(30, 250),
    alpha=0.75,
    palette='tab10',
    ax=ax
)
ax.set_title('Revenue vs Spend — hue=channel, size=budget, style=region')
ax.set_xlabel('Ad Spend ($)')
ax.set_ylabel('Revenue ($)')
ax.legend(bbox_to_anchor=(1.01, 1), loc='upper left', fontsize=8)
plt.tight_layout()
plt.savefig('scatter_hue_size_style.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(33)
sns.set_theme(style='whitegrid', font_scale=0.9)

# Synthetic multi-channel, multi-region dataset
channels = ['Search', 'Social', 'Email', 'Display', 'Affiliate']
regions  = ['North', 'South', 'East']
rows = []
for ch in channels:
    slope = {'Search': 4.0, 'Social': 2.5, 'Email': 5.5,
             'Display': 1.5, 'Affiliate': 3.2}[ch]
    for region in regions:
        n = 30
        spend = np.random.uniform(200, 8000, n)
        rev   = spend * slope + np.random.randn(n) * spend * 0.25
        for s, r in zip(spend, rev):
            rows.append({'channel': ch, 'region': region,
                         'spend': round(s, 0), 'revenue': round(r, 0)})
df = pd.DataFrame(rows)

# relplot with col=channel, col_wrap=3, hue+style=region
# Each hue level gets a distinct marker AND color automatically
g = sns.relplot(
    data=df, x='spend', y='revenue',
    col='channel', col_wrap=3,
    hue='region', style='region',     # different marker per region
    markers=['o', 's', '^'],          # explicit marker list
    palette='Set2',
    alpha=0.65, s=40,
    height=3, aspect=1.2,
    kind='scatter'
)
g.set_titles('{col_name}')
g.set_axis_labels('Ad Spend ($)', 'Revenue ($)')
g.figure.suptitle('relplot: Spend vs Revenue — col_wrap=3, marker per region', y=1.03)
g.add_legend(title='Region')
plt.tight_layout()
plt.savefig('scatter_relplot_colwrap_markers.png', dpi=80)
plt.close()
print("relplot col_wrap + marker styles saved.")

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd

sns.set_theme(style='white')
df = sns.load_dataset('tips')

# Select numeric columns
corr = df[['total_bill','tip','size']].corr()

fig, ax = plt.subplots(figsize=(5, 4))
sns.heatmap(corr, annot=True, fmt='.2f', cmap='RdBu_r',
            vmin=-1, vmax=1, linewidths=0.5,
            square=True, ax=ax)
ax.set_title('Feature Correlation Matrix')
plt.tight_layout()
plt.savefig('heatmap_corr.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(9)
months = ['Jan','Feb','Mar','Apr','May','Jun']
regions= ['North','South','East','West']
data   = pd.DataFrame({
    'month':   np.tile(months, 4),
    'region':  np.repeat(regions, 6),
    'revenue': np.random.uniform(50, 200, 24).round(1)
})

pivot = data.pivot(index='region', columns='month', values='revenue')

fig, ax = plt.subplots(figsize=(9, 4))
sns.heatmap(pivot, annot=True, fmt='.0f', cmap='YlGnBu',
            linewidths=0.4, ax=ax, cbar_kws=dict(label='Revenue ($K)'))
ax.set_title('Monthly Revenue by Region ($K)')
ax.set_xlabel(''); ax.set_ylabel('')
plt.tight_layout()
plt.savefig('heatmap_pivot.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
# Simulate a 4-class confusion matrix
classes = ['Cat', 'Dog', 'Bird', 'Fish']
# True labels and predicted labels
true_labels = np.random.choice(classes, 200, p=[0.3, 0.3, 0.2, 0.2])
# Add some misclassification noise
pred_labels = true_labels.copy()
noise_idx = np.random.choice(len(true_labels), 40, replace=False)
pred_labels[noise_idx] = np.random.choice(classes, 40)

# Build confusion matrix manually
cm = pd.crosstab(pd.Series(true_labels, name='Actual'),
                 pd.Series(pred_labels, name='Predicted'))
# Ensure all classes present
cm = cm.reindex(index=classes, columns=classes, fill_value=0)

fig, axes = plt.subplots(1, 2, figsize=(12, 4))

# Raw counts
sns.heatmap(cm, annot=True, fmt='d', cmap='Blues',
            linewidths=0.5, ax=axes[0])
axes[0].set_title('Confusion Matrix (counts)')

# Normalized (recall per class)
cm_norm = cm.div(cm.sum(axis=1), axis=0).round(2)
sns.heatmap(cm_norm, annot=True, fmt='.2f', cmap='YlOrRd',
            vmin=0, vmax=1, linewidths=0.5, ax=axes[1])
axes[1].set_title('Confusion Matrix (row-normalized recall)')

plt.tight_layout()
plt.savefig('heatmap_confusion.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
sns.set_theme(style='white')

# --- Part 1: lower-triangle heatmap using the mask parameter ---
iris = sns.load_dataset('iris')
corr = iris[['sepal_length','sepal_width','petal_length','petal_width']].corr()

# mask=True hides that cell; np.triu masks upper triangle (keep lower + diagonal)
mask_upper = np.triu(np.ones_like(corr, dtype=bool), k=1)

fig, ax = plt.subplots(figsize=(5, 4))
sns.heatmap(corr, mask=mask_upper, annot=True, fmt='.2f',
            cmap='coolwarm', vmin=-1, vmax=1,
            square=True, linewidths=0.5,
            cbar_kws=dict(shrink=0.7), ax=ax)
ax.set_title('Iris Correlation — lower triangle only (mask=triu)')
plt.tight_layout()
plt.savefig('heatmap_triangle_mask.png', dpi=80)
plt.close()

# --- Part 2: clustermap — hierarchically clusters rows AND columns ---
np.random.seed(7)
n_genes, n_samples = 20, 12
data = pd.DataFrame(
    np.random.randn(n_genes, n_samples),
    index=[f'Gene_{i:02d}' for i in range(n_genes)],
    columns=[f'S{j:02d}' for j in range(n_samples)]
)
# Add block structure so clustering is visible
data.iloc[:8,  :6]  += 2   # high block top-left
data.iloc[12:, 6:]  -= 2   # low block bottom-right

g = sns.clustermap(
    data,
    cmap='RdBu_r', center=0,
    figsize=(9, 7),
    dendrogram_ratio=(0.12, 0.12),
    cbar_pos=(0.02, 0.85, 0.03, 0.12),
    linewidths=0.3,
    method='ward'
)
g.figure.suptitle('clustermap: Hierarchical Clustering of Gene Expression', y=1.01)
plt.savefig('heatmap_clustermap.png', dpi=80, bbox_inches='tight')
plt.close()
print("Triangle mask heatmap and clustermap saved.")

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

g = sns.pairplot(
    iris, hue='species',
    palette='Set2',
    plot_kws=dict(alpha=0.5, s=25),
    diag_kind='kde'
)
g.figure.suptitle('Iris Dataset — Pair Plot', y=1.02)
plt.tight_layout()
plt.savefig('pairplot_iris.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')
cols = ['total_bill', 'tip', 'size']

g = sns.pairplot(
    tips[cols + ['time']], hue='time',
    kind='reg',          # scatter + regression line
    diag_kind='hist',    # histogram on diagonal
    palette='Set1',
    plot_kws=dict(scatter_kws=dict(alpha=0.3, s=20),
                  line_kws=dict(linewidth=1.5))
)
g.figure.suptitle('Tips Dataset — Regression Pair Plot', y=1.02)
plt.tight_layout()
plt.savefig('pairplot_reg.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')
cols = ['sepal_length', 'sepal_width', 'petal_length']

g = sns.PairGrid(iris[cols + ['species']], hue='species',
                 palette='tab10', diag_sharey=False)

# Upper triangle: scatter
g.map_upper(sns.scatterplot, alpha=0.4, s=20)

# Lower triangle: KDE contours
g.map_lower(sns.kdeplot, fill=True, alpha=0.25, levels=4)

# Diagonal: histogram
g.map_diag(sns.histplot, kde=True, alpha=0.5)

g.add_legend()
g.figure.suptitle('PairGrid: Custom Upper/Lower/Diagonal', y=1.02)
plt.tight_layout()
plt.savefig('pairgrid_custom.png', dpi=80)
plt.close()

import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np, pandas as pd
from scipy import stats

sns.set_theme(style='whitegrid', font_scale=1.1)
rng = np.random.default_rng(7)

n = 200
df = pd.DataFrame({
    'market_cap':   rng.lognormal(8, 1.5, n),
    'pe_ratio':     rng.lognormal(3, 0.5, n),
    'debt_equity':  rng.exponential(1.2, n),
    'revenue_growth': rng.normal(0.15, 0.12, n),
    'return_1y':    rng.normal(0.08, 0.22, n),
})
# Add mild relationship
df['return_1y'] += 0.03 * np.log(df['market_cap']) - 0.02 * df['pe_ratio'] + rng.normal(0, 0.05, n)

fig, axes = plt.subplots(2, 2, figsize=(12, 10))

# 1. Residual plot
slope, intercept, r, p, se = stats.linregress(df['pe_ratio'], df['return_1y'])
residuals = df['return_1y'] - (slope * df['pe_ratio'] + intercept)
axes[0,0].scatter(df['pe_ratio'], residuals, alpha=0.5, color='steelblue', s=25)
axes[0,0].axhline(0, color='red', ls='--', lw=1.5)
axes[0,0].set(title='Residual Plot (PE vs Return)', xlabel='P/E Ratio', ylabel='Residual')
axes[0,0].text(0.05, 0.95, f'r={r:.3f}', transform=axes[0,0].transAxes,
               va='top', fontsize=10)

# 2. Q-Q plot for normality
(osm, osr), (slope2, intercept2, r2) = stats.probplot(df['return_1y'])
axes[0,1].scatter(osm, osr, alpha=0.5, color='coral', s=20)
axes[0,1].plot(osm, slope2*np.array(osm)+intercept2, 'r-', lw=2)
axes[0,1].set(title='Q-Q Plot: Return (1Y)', xlabel='Theoretical Quantiles', ylabel='Sample Quantiles')

# 3. Log transform effect
axes[1,0].hist(df['market_cap'], bins=30, color='seagreen', alpha=0.7, edgecolor='white')
axes[1,0].set(title='Market Cap (Raw)', xlabel='Market Cap')

axes[1,1].hist(np.log(df['market_cap']), bins=30, color='seagreen', alpha=0.7, edgecolor='white')
axes[1,1].set(title='Market Cap (Log-Transformed)', xlabel='log(Market Cap)')

plt.suptitle('Financial Feature Diagnostics', fontsize=14, y=1.01)
plt.tight_layout()
plt.savefig('financial_diagnostics.png', dpi=100)
plt.show()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

g = sns.FacetGrid(tips, col='time', row='smoker',
                  height=3, aspect=1.2, margin_titles=True)
g.map_dataframe(sns.histplot, x='total_bill', bins=15, kde=True)
g.set_axis_labels('Total Bill ($)', 'Count')
g.set_titles(row_template='{row_name}', col_template='{col_name}')
g.figure.suptitle('Total Bill by Time & Smoker', y=1.03)
plt.tight_layout()
plt.savefig('facet_grid.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

# catplot wraps barplot/boxplot/etc into a FacetGrid
g = sns.catplot(
    data=tips, x='day', y='tip',
    col='time', kind='box',
    palette='Set2',
    height=4, aspect=0.9
)
g.set_titles('{col_name}')
g.set_axis_labels('Day', 'Tip ($)')
g.figure.suptitle('Tip Distribution by Day & Meal Time', y=1.03)
plt.tight_layout()
plt.savefig('facet_catplot.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(5)
sns.set_theme(style='whitegrid', font_scale=0.85)

# Synthetic multi-group data
groups = ['A', 'B', 'C']
conditions = ['Low', 'High']
rows = []
for grp in groups:
    for cond in conditions:
        n = 40
        x = np.random.uniform(0, 100, n)
        slope = {'A': 1.5, 'B': 0.8, 'C': 2.2}[grp]
        offset = {'Low': 0, 'High': 50}[cond]
        y = x * slope + offset + np.random.randn(n) * 20
        rows.extend({'group': grp, 'condition': cond,
                     'x': round(xi, 1), 'y': round(yi, 1)}
                    for xi, yi in zip(x, y))
df = pd.DataFrame(rows)

g = sns.FacetGrid(df, col='group', row='condition',
                  height=3, aspect=1.2,
                  margin_titles=True, sharey=False)
g.map_dataframe(sns.scatterplot, x='x', y='y', alpha=0.5, s=20)
g.map_dataframe(sns.regplot, x='x', y='y',
                scatter=False,
                line_kws=dict(color='red', linewidth=1.5))
g.set_axis_labels('X', 'Y')
g.set_titles(col_template='{col_name}', row_template='{row_name}')
g.figure.suptitle('Faceted Scatter with Regression', y=1.03)
plt.tight_layout()
plt.savefig('facet_scatter_reg.png', dpi=80)
plt.close()

import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np, pandas as pd
from scipy import stats

sns.set_theme(style='white', font_scale=1.1)
rng = np.random.default_rng(42)

regions  = ['North','South','East','West','Central']
products = ['Widget','Gadget','Gizmo','Doohickey']
months   = ['Jan','Feb','Mar','Apr','May','Jun']

# Simulate monthly revenue per region/product
data = rng.uniform(50, 300, (len(regions), len(months)))
df = pd.DataFrame(data, index=regions, columns=months)

# Compute growth vs prior month
growth = df.pct_change(axis=1) * 100
growth.iloc[:, 0] = rng.uniform(-5, 20, len(regions))  # fill first month

# Significance test: is growth > 0?
sig_markers = pd.DataFrame('', index=regions, columns=months)
for region in regions:
    for month in months:
        g = growth.loc[region, month]
        sig_markers.loc[region, month] = '★' if g > 15 else ('▼' if g < -5 else '')

# Plot
fig, axes = plt.subplots(1, 2, figsize=(14, 5))

sns.heatmap(df, annot=True, fmt='.0f', cmap='YlOrRd',
            linewidths=0.5, ax=axes[0], cbar_kws={'label': 'Revenue ($K)'})
axes[0].set_title('Monthly Revenue by Region', fontsize=12)

# Growth heatmap with significance markers
annot = growth.round(1).astype(str) + '%
' + sig_markers
sns.heatmap(growth, annot=annot, fmt='', cmap='RdYlGn', center=0,
            linewidths=0.5, ax=axes[1], cbar_kws={'label': 'MoM Growth (%)'},
            annot_kws={'size': 9})
axes[1].set_title('Month-over-Month Growth
★=High Growth ▼=Decline', fontsize=12)

plt.suptitle('Regional Sales Performance Analysis', fontsize=13, y=1.02)
plt.tight_layout()
plt.savefig('grouped_heatmap.png', dpi=100)
plt.show()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(5)
weeks   = list(range(1, 13))
regions = ['North','South','East']

rows = []
for region in regions:
    base = np.random.uniform(80, 120)
    trend= np.random.uniform(1, 5)
    for w in weeks:
        for _ in range(5):   # 5 reps per point → CI makes sense
            rows.append({
                'week':   w,
                'region': region,
                'sales':  base + trend * w + np.random.randn() * 15
            })
df = pd.DataFrame(rows)

sns.set_theme(style='whitegrid')
fig, ax = plt.subplots(figsize=(10, 4))
sns.lineplot(data=df, x='week', y='sales', hue='region',
             palette='Set2', linewidth=2.5, ax=ax)
ax.set_title('Weekly Sales by Region (with 95% CI)')
ax.set_xlabel('Week'); ax.set_ylabel('Sales ($K)')
plt.tight_layout()
plt.savefig('line_hue_ci.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(9)
products = ['Widget','Gadget','Doohickey']
months   = pd.date_range('2024-01-01', periods=12, freq='MS')

rows = []
for prod in products:
    base = np.random.uniform(50, 200)
    vals = base + np.cumsum(np.random.randn(12) * 10)
    for m, v in zip(months, vals):
        rows.append({'month':m,'product':prod,'revenue':max(v,10)})
df = pd.DataFrame(rows)

g = sns.relplot(data=df, x='month', y='revenue',
                col='product', kind='line',
                height=3, aspect=1.3,
                marker='o', markersize=5)
g.set_titles('{col_name}')
for ax in g.axes.flat:
    ax.xaxis.set_major_formatter(plt.matplotlib.dates.DateFormatter('%b'))
    ax.tick_params(axis='x', rotation=45)
g.figure.suptitle('Monthly Revenue by Product', y=1.03)
plt.tight_layout()
plt.savefig('line_relplot.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(15)
sns.set_theme(style='whitegrid')

months = list(range(1, 13))
products = ['Alpha', 'Beta', 'Gamma']
rows = []
for prod in products:
    base = np.random.uniform(100, 300)
    vals = base + np.cumsum(np.random.randn(12) * 20)
    for m, v in zip(months, vals):
        rows.append({'month': m, 'product': prod, 'revenue': max(v, 10)})
df = pd.DataFrame(rows)

fig, ax = plt.subplots(figsize=(11, 4))
sns.lineplot(data=df, x='month', y='revenue', hue='product',
             marker='o', markersize=6, linewidth=2,
             palette='tab10', ax=ax)

# Shade a promotion period
ax.axvspan(5, 7, alpha=0.12, color='green', label='Promo period')

# Annotate peak revenue
peak = df.loc[df['revenue'].idxmax()]
ax.annotate(f"Peak: {peak['product']}
${peak['revenue']:.0f}K",
            xy=(peak['month'], peak['revenue']),
            xytext=(peak['month'] + 0.5, peak['revenue'] - 30),
            arrowprops=dict(arrowstyle='->', color='red'),
            fontsize=8, color='red')

ax.set_title('Monthly Revenue with Event Shading')
ax.set_xlabel('Month'); ax.set_ylabel('Revenue ($K)')
ax.set_xticks(months)
ax.set_xticklabels(['Jan','Feb','Mar','Apr','May','Jun',
                    'Jul','Aug','Sep','Oct','Nov','Dec'], rotation=30)
ax.legend(loc='upper left', fontsize=8)
plt.tight_layout()
plt.savefig('line_annotated.png', dpi=80)
plt.close()

import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np, pandas as pd

sns.set_theme(style='darkgrid', font_scale=1.0)
rng = np.random.default_rng(42)

n = 168  # 1 week of hourly data
t = pd.date_range('2024-01-01', periods=n, freq='h')
cpu    = 40 + 20*np.sin(np.linspace(0, 4*np.pi, n)) + rng.normal(0, 5, n)
memory = 60 + 10*np.sin(np.linspace(0, 2*np.pi, n)) + rng.normal(0, 3, n)
# Inject anomalies
cpu[[24, 72, 120, 145]] += rng.uniform(35, 50, 4)
memory[[36, 96, 130]]    += rng.uniform(25, 35, 3)

df = pd.DataFrame({'time': t, 'CPU': cpu.clip(0,100), 'Memory': memory.clip(0,100)})

# Detect anomalies via z-score
for col in ['CPU', 'Memory']:
    mu, sigma = df[col].mean(), df[col].std()
    df[f'{col}_anomaly'] = (df[col] - mu).abs() > 2.5 * sigma

fig, axes = plt.subplots(2, 1, figsize=(14, 8), sharex=True)

for ax, metric, color in zip(axes, ['CPU', 'Memory'], ['steelblue', 'seagreen']):
    sns.lineplot(data=df, x='time', y=metric, ax=ax, color=color, lw=1.5, label=metric)
    # Shade anomaly regions
    anomalies = df[df[f'{metric}_anomaly']]
    ax.scatter(anomalies['time'], anomalies[metric],
               color='red', s=60, zorder=5, label='Anomaly')
    ax.axhline(df[metric].mean() + 2.5*df[metric].std(),
               color='red', ls='--', lw=1, alpha=0.6, label='2.5sigma threshold')
    ax.set(ylabel=f'{metric} Usage (%)', title=f'{metric} Usage — Anomaly Detection')
    ax.legend(loc='upper right')

axes[-1].set_xlabel('Time')
plt.suptitle('Server Metrics Anomaly Dashboard', fontsize=13, y=1.01)
plt.tight_layout()
plt.savefig('anomaly_dashboard.png', dpi=100)
plt.show()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import matplotlib.ticker as mticker

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, ax = plt.subplots(figsize=(8, 4))
sns.boxplot(data=tips, x='day', y='total_bill', palette='pastel', ax=ax)

# Matplotlib customizations on top
ax.set_title('Total Bill by Day', fontsize=14, fontweight='bold', pad=12)
ax.set_xlabel('')
ax.set_ylabel('Total Bill', fontsize=11)
ax.yaxis.set_major_formatter(mticker.StrMethodFormatter('${x:.0f}'))
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)

# Add median annotations
medians = tips.groupby('day')['total_bill'].median()
for i, (day, med) in enumerate(medians.items()):
    ax.text(i, med + 0.5, f'${med:.0f}', ha='center', fontsize=9, color='darkred')

plt.tight_layout()
plt.savefig('custom_axes.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

fig, ax = plt.subplots(figsize=(7, 5))

# Layer 1: violin
sns.violinplot(data=iris, x='species', y='petal_length',
               palette='pastel', inner=None, ax=ax)
# Layer 2: box inside violin
sns.boxplot(data=iris, x='species', y='petal_length',
            width=0.15, fliersize=0,
            boxprops=dict(facecolor='white', zorder=2), ax=ax)
# Layer 3: points
sns.stripplot(data=iris, x='species', y='petal_length',
              color='black', size=2.5, alpha=0.4, jitter=True, ax=ax)

ax.set_title('Petal Length by Species')
ax.set_ylabel('Petal Length (cm)')
plt.tight_layout()
plt.savefig('custom_layered.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import matplotlib.ticker as mticker
import numpy as np
import pandas as pd

np.random.seed(42)
sns.set_theme(style='ticks', palette='colorblind', font_scale=1.0)

# Synthetic dataset
df = pd.DataFrame({
    'method': np.repeat(['Baseline', 'Model A', 'Model B', 'Model C'], 50),
    'accuracy': np.concatenate([
        np.random.normal(0.72, 0.04, 50),
        np.random.normal(0.81, 0.03, 50),
        np.random.normal(0.78, 0.05, 50),
        np.random.normal(0.85, 0.03, 50),
    ])
})

fig, ax = plt.subplots(figsize=(8, 4))
order = df.groupby('method')['accuracy'].mean().sort_values().index

sns.violinplot(data=df, x='method', y='accuracy', order=order,
               palette='colorblind', inner='box', ax=ax)

ax.yaxis.set_major_formatter(mticker.PercentFormatter(xmax=1, decimals=0))
ax.set_xlabel(''); ax.set_ylabel('Accuracy')
ax.set_title('Model Accuracy Comparison', fontsize=13, fontweight='bold')
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)

# Significance bracket
best_mean  = df[df['method']=='Model C']['accuracy'].mean()
base_mean  = df[df['method']=='Baseline']['accuracy'].mean()
ax.annotate(f'+{(best_mean-base_mean)*100:.1f}% vs Baseline',
            xy=(0.5, 0.96), xycoords='axes fraction',
            ha='center', fontsize=9, color='darkgreen',
            fontweight='bold')

plt.tight_layout()
plt.savefig('custom_publication.png', dpi=80)
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np, pandas as pd

rng = np.random.default_rng(7)
tips = sns.load_dataset('tips')

# Default style
fig, axes = plt.subplots(1, 3, figsize=(14, 4))

sns.histplot(tips['total_bill'], kde=True, ax=axes[0])
axes[0].set_title('Default Style')

# Paper context — smaller for publications
with sns.plotting_context('paper', font_scale=1.2):
    sns.histplot(tips['total_bill'], kde=True, color='seagreen', ax=axes[1])
    axes[1].set_title('Paper Context')

# Talk context — larger for presentations
with sns.plotting_context('talk', font_scale=0.9):
    sns.histplot(tips['tip'], kde=True, color='coral', ax=axes[2])
    axes[2].set_title('Talk Context')

plt.suptitle('Seaborn Context Comparison', fontsize=12)
plt.tight_layout()
plt.savefig('context_comparison.png', dpi=100)
plt.close()

# Override with rc_context — dark background for a single plot
with sns.axes_style('darkgrid'):
    with plt.rc_context({'figure.facecolor': '#1e1e2e',
                         'axes.facecolor':   '#1e1e2e',
                         'axes.labelcolor':  'white',
                         'xtick.color':      'white',
                         'ytick.color':      'white',
                         'text.color':       'white'}):
        fig2, ax2 = plt.subplots(figsize=(6, 4))
        sns.scatterplot(data=tips, x='total_bill', y='tip',
                        hue='time', palette=['#ffa600','#58508d'],
                        alpha=0.8, s=60, ax=ax2)
        ax2.set_title('Dark Theme Override', color='white', fontsize=13)
        ax2.legend(labelcolor='white', facecolor='#2e2e3e')
        plt.tight_layout()
        plt.savefig('dark_theme.png', dpi=100, facecolor='#1e1e2e')
        plt.close()
print("Context comparison and dark theme override saved.")

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')
g = sns.FacetGrid(tips, col='time', hue='smoker', height=4, aspect=0.8)
g.map(sns.scatterplot, 'total_bill', 'tip', alpha=0.7)
g.add_legend(); g.set_axis_labels('Total Bill ($)', 'Tip ($)')
g.set_titles(col_template='{col_name}')
g.figure.suptitle('Tips by Time and Smoker Status', y=1.02)
plt.savefig('facetgrid_tips.png', dpi=80, bbox_inches='tight'); plt.close()
print('Saved facetgrid_tips.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

iris = sns.load_dataset('iris')
g = sns.PairGrid(iris, hue='species')
g.map_diag(sns.histplot, alpha=0.6)
g.map_upper(sns.scatterplot, alpha=0.6)
g.map_lower(sns.kdeplot)
g.add_legend()
plt.savefig('pairgrid_iris.png', dpi=80, bbox_inches='tight'); plt.close()
print('Saved pairgrid_iris.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')
g = sns.catplot(data=tips, x='day', y='total_bill',
                col='time', hue='sex',
                kind='box', height=4, aspect=0.8)
g.set_axis_labels('Day', 'Total Bill ($)')
g.set_titles(col_template='{col_name}')
g.figure.suptitle('Bills by Day, Time, and Sex', y=1.02)
plt.savefig('catplot_tips.png', dpi=80, bbox_inches='tight'); plt.close()
print('Saved catplot_tips.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

fmri = sns.load_dataset('fmri')
g = sns.relplot(data=fmri, x='timepoint', y='signal',
               col='region', hue='event',
               kind='line', height=4, aspect=0.9,
               errorbar='se')
g.set_titles(col_template='Region: {col_name}')
g.figure.suptitle('fMRI Signal by Region and Event', y=1.02)
plt.savefig('relplot_fmri.png', dpi=80, bbox_inches='tight'); plt.close()
print('Saved relplot_fmri.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')
fig, ax = plt.subplots(figsize=(8, 5))
sns.barplot(data=tips, x='day', y='total_bill', hue='sex',
            errorbar='ci', capsize=0.1, alpha=0.85, ax=ax)
ax.set_title('Mean Total Bill with 95% CI')
ax.set_xlabel('Day'); ax.set_ylabel('Total Bill ($)')
plt.tight_layout()
plt.savefig('barplot_ci.png', dpi=80); plt.close()
print('Saved barplot_ci.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
from scipy import stats

np.random.seed(42)
group_a = np.random.normal(42, 8, 50)
group_b = np.random.normal(48, 9, 50)
import pandas as pd
df = pd.DataFrame({'value': np.concatenate([group_a, group_b]), 'group': ['A']*50+['B']*50})

fig, ax = plt.subplots(figsize=(6, 5))
sns.boxplot(data=df, x='group', y='value', ax=ax, palette='Set2')

_, p = stats.ttest_ind(group_a, group_b)
stars = '***' if p < 0.001 else '**' if p < 0.01 else '*' if p < 0.05 else 'ns'
y_max = df['value'].max() + 2
ax.plot([0, 0, 1, 1], [y_max, y_max+1, y_max+1, y_max], lw=1.5, color='black')
ax.text(0.5, y_max+1.2, stars, ha='center', va='bottom', fontsize=14)
ax.set_title(f'Group Comparison (p={p:.4f})')
plt.tight_layout()
plt.savefig('significance.png', dpi=80); plt.close()
print(f'Saved significance.png ({stars})')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')
fig, axes = plt.subplots(1, 2, figsize=(12, 5))

sns.pointplot(data=tips, x='day', y='total_bill', hue='sex',
              errorbar='sd', capsize=0.1, dodge=True, ax=axes[0])
axes[0].set_title('Pointplot with SD error bars')

sns.pointplot(data=tips, x='day', y='tip', hue='smoker',
              errorbar='se', join=False, dodge=0.4, ax=axes[1])
axes[1].set_title('Pointplot (no join) with SE')
plt.tight_layout()
plt.savefig('pointplot.png', dpi=80); plt.close()
print('Saved pointplot.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from scipy import stats
from itertools import combinations

np.random.seed(42)
groups = {'Control': np.random.normal(10,2,40), 'Drug A': np.random.normal(12,2,40), 'Drug B': np.random.normal(14,3,40)}
df = pd.DataFrame([(v, g) for g, vals in groups.items() for v in vals], columns=['response','group'])

fig, ax = plt.subplots(figsize=(7, 6))
sns.boxplot(data=df, x='group', y='response', ax=ax, palette='pastel')

pairs = list(combinations(groups.keys(), 2))
y_max = df['response'].max()
for i, (g1, g2) in enumerate(pairs):
    _, p = stats.ttest_ind(groups[g1], groups[g2])
    stars = '***' if p<0.001 else '**' if p<0.01 else '*' if p<0.05 else 'ns'
    x1 = list(groups.keys()).index(g1)
    x2 = list(groups.keys()).index(g2)
    y = y_max + 1.5*(i+1)
    ax.plot([x1, x1, x2, x2], [y, y+0.3, y+0.3, y], lw=1.2, color='black')
    ax.text((x1+x2)/2, y+0.4, stars, ha='center', fontsize=12)
ax.set_title('Multi-Group Significance Test')
plt.tight_layout()
plt.savefig('multi_sig.png', dpi=80); plt.close()
print('Saved multi_sig.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')
g = sns.lmplot(data=tips, x='total_bill', y='tip', hue='smoker',
               ci=95, scatter_kws={'alpha':0.5}, height=5, aspect=1.2)
g.set_axis_labels('Total Bill ($)', 'Tip ($)')
g.axes[0,0].set_title('Tip vs Bill by Smoker Status (95% CI)')
plt.savefig('lmplot.png', dpi=80, bbox_inches='tight'); plt.close()
print('Saved lmplot.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np

tips = sns.load_dataset('tips')
fig, axes = plt.subplots(1, 2, figsize=(12, 5))
sns.residplot(data=tips, x='total_bill', y='tip', lowess=True, ax=axes[0],
              scatter_kws={'alpha':0.5}, line_kws={'color':'red','linewidth':2})
axes[0].set_title('Residual Plot (lowess trend)')
axes[0].axhline(0, color='gray', linestyle='--')

sns.residplot(data=tips, x='size', y='total_bill', ax=axes[1],
              scatter_kws={'alpha':0.5})
axes[1].set_title('Residual Plot: size vs bill')
axes[1].axhline(0, color='gray', linestyle='--')
plt.tight_layout()
plt.savefig('residplot.png', dpi=80); plt.close()
print('Saved residplot.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')
fig, axes = plt.subplots(1, 3, figsize=(15, 5))
for ax, kind in zip(axes, ['scatter', 'kde', 'hex']):
    g = sns.jointplot(data=tips, x='total_bill', y='tip', kind=kind, height=4)
    g.set_axis_labels('Total Bill ($)', 'Tip ($)')
    g.figure.suptitle(f'Joint Plot: {kind}', y=1.02)
    g.savefig(f'joint_{kind}.png', dpi=80)
    plt.close(g.figure)
    print(f'Saved joint_{kind}.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np

tips = sns.load_dataset('tips')
fig, axes = plt.subplots(1, 2, figsize=(12, 5))

# Polynomial regression
sns.regplot(data=tips, x='total_bill', y='tip', order=2, ax=axes[0],
            scatter_kws={'alpha':0.4}, line_kws={'color':'red','linewidth':2})
axes[0].set_title('Polynomial Regression (degree=2)')

# ECDF for comparing distributions
for day in ['Thur','Fri','Sat','Sun']:
    subset = tips[tips['day']==day]
    sns.ecdfplot(data=subset, x='total_bill', ax=axes[1], label=day)
axes[1].set_title('ECDF of Total Bill by Day')
axes[1].legend(); axes[1].set_xlabel('Total Bill ($)')
plt.tight_layout()
plt.savefig('regplot_ecdf.png', dpi=80); plt.close()
print('Saved regplot_ecdf.png')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
df = pd.DataFrame({'value': np.random.randn(300), 'group': np.repeat(['A','B','C'], 100)})
g = sns.FacetGrid(df, col='group', height=3.5, aspect=0.9)
g.map(sns.histplot, 'value', kde=True, bins=20)
g.set_titles(col_template='{col_name}')
g.figure.suptitle('FacetGrid by Group', y=1.02)
g.savefig('facetgrid_hist.png', dpi=100, bbox_inches='tight')
print('Saved facetgrid_hist.png')
plt.close('all')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

tips = sns.load_dataset('tips')
g = sns.FacetGrid(tips, row='sex', col='time', height=3, aspect=1.2, margin_titles=True)
g.map(sns.scatterplot, 'total_bill', 'tip', alpha=0.6)
g.add_legend()
g.savefig('facetgrid_2d.png', dpi=100, bbox_inches='tight')
print('Saved facetgrid_2d.png')
plt.close('all')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

iris = sns.load_dataset('iris')
g = sns.PairGrid(iris, hue='species', vars=['sepal_length','petal_length','petal_width'])
g.map_upper(sns.scatterplot, alpha=0.6)
g.map_lower(sns.kdeplot, fill=True, alpha=0.4)
g.map_diag(sns.histplot, kde=True)
g.add_legend()
g.savefig('pairgrid_mixed.png', dpi=80, bbox_inches='tight')
print('Saved pairgrid_mixed.png')
plt.close('all')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

def scatter_means(x, y, **kw):
    plt.scatter(x, y, alpha=0.5, **{k:v for k,v in kw.items() if k!='label'})
    plt.axhline(y.mean(), color='red', ls='--', lw=1.5)
    plt.axvline(x.mean(), color='blue', ls='--', lw=1.5)

np.random.seed(42)
df = pd.DataFrame({'x':np.random.randn(150),'y':np.random.randn(150),
                   'group':np.repeat(['G1','G2','G3'],50)})
g = sns.FacetGrid(df, col='group', height=3)
g.map(scatter_means, 'x', 'y')
g.figure.suptitle('Scatter with Group Means', y=1.02)
g.savefig('facetgrid_custom.png', dpi=100, bbox_inches='tight')
print('Saved facetgrid_custom.png')
plt.close('all')

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

np.random.seed(42)
df = pd.DataFrame({
    'score': np.concatenate([np.random.normal(70,10,80), np.random.normal(75,8,80), np.random.normal(65,12,80)]),
    'class': ['A']*80 + ['B']*80 + ['C']*80
})
fig, ax = plt.subplots(figsize=(7, 5))
sns.violinplot(data=df, x='class', y='score', inner=None, palette='muted', alpha=0.7, ax=ax)
sns.swarmplot(data=df, x='class', y='score', color='black', size=2.5, alpha=0.6, ax=ax)
ax.set_title('Score Distribution by Class (Violin + Swarm)')
fig.savefig('violin_swarm.png', dpi=100, bbox_inches='tight')
print('Saved violin_swarm.png')
plt.close()

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
df = pd.DataFrame({
    'response_ms': np.concatenate([np.random.exponential(200,300),
                                   np.random.exponential(150,300),
                                   np.random.exponential(300,300)]),
    'server': ['A']*300 + ['B']*300 + ['C']*300
})
fig, ax = plt.subplots(figsize=(8, 5))
sns.ecdfplot(data=df, x='response_ms', hue='server', ax=ax)
ax.axvline(200, color='gray', ls=':', label='200ms target')
ax.set_title('ECDF: Response Time by Server')
ax.set_xlabel('Response Time (ms)')
ax.legend()
fig.savefig('ecdf_comparison.png', dpi=100, bbox_inches='tight')
print('Saved ecdf_comparison.png')
plt.close()

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np

np.random.seed(42)
x = np.linspace(0, 10, 100)
y = 2*x + np.random.normal(0, 2, 100)
fig, axes = plt.subplots(1, 2, figsize=(11, 4))
sns.regplot(x=x, y=y, ax=axes[0])
axes[0].set_title('Regression with CI')
sns.residplot(x=x, y=y, ax=axes[1])
axes[1].axhline(0, color='red', ls='--')
axes[1].set_title('Residuals vs Fitted')
fig.tight_layout()
fig.savefig('residual_diagnostic.png', dpi=100, bbox_inches='tight')
print('Saved residual_diagnostic.png')
plt.close()

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
df = pd.DataFrame({
    'value': np.concatenate([np.random.normal(5,1,50), np.random.normal(7,1.5,50), np.random.normal(6.2,1.2,50)]),
    'group': ['Control']*50 + ['Drug A']*50 + ['Drug B']*50
})
fig, ax = plt.subplots(figsize=(7, 5))
sns.boxplot(data=df, x='group', y='value', palette='Set2', ax=ax)
y_max = df['value'].max() + 0.5
ax.plot([0, 1], [y_max, y_max], 'k-', lw=1.5)
ax.text(0.5, y_max+0.1, '***', ha='center', fontsize=14)
ax.set_title('Drug Effect with Significance Bracket')
fig.savefig('boxplot_sig.png', dpi=100, bbox_inches='tight')
print('Saved boxplot_sig.png')
plt.close()

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
df = pd.DataFrame({'x':np.random.randn(100),'y':np.random.randn(100)})
fig, axes = plt.subplots(2, 2, figsize=(12, 9))
for ax, ctx in zip(axes.flat, ['paper','notebook','talk','poster']):
    with sns.plotting_context(ctx):
        sns.scatterplot(data=df, x='x', y='y', ax=ax, alpha=0.6)
        ax.set_title(f'Context: {ctx}')
fig.suptitle('Seaborn Contexts', fontsize=14)
fig.tight_layout()
fig.savefig('contexts.png', dpi=100, bbox_inches='tight')
print('Saved contexts.png')
plt.close()

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='darkgrid', palette='bright', font_scale=1.1,
              rc={'axes.facecolor':'#1e1e2e','figure.facecolor':'#1e1e2e',
                  'text.color':'white','axes.labelcolor':'white',
                  'xtick.color':'white','ytick.color':'white'})
np.random.seed(42)
df = pd.DataFrame({'x':np.random.randn(200),'y':np.random.randn(200),'g':np.random.choice(['A','B','C'],200)})
fig, ax = plt.subplots(figsize=(7,5))
sns.scatterplot(data=df, x='x', y='y', hue='g', alpha=0.7, ax=ax)
ax.set_title('Dark Mode Plot')
fig.savefig('dark_theme.png', dpi=100, bbox_inches='tight')
print('Saved dark_theme.png')
sns.set_theme()
plt.close()

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt

fig, axes = plt.subplots(2, 2, figsize=(12, 8))
tips = sns.load_dataset('tips')
pals = [('Blues',4), ('husl',4), ('Set1',4), (['#FF6B6B','#4ECDC4','#45B7D1','#96CEB4'],4)]
titles = ['Blues (seq)','HUSL (qual)','Set1 (qual)','Custom hex']
for ax, (pal, _), title in zip(axes.flat, pals, titles):
    with sns.axes_style('whitegrid'):
        sns.boxplot(data=tips, x='day', y='total_bill', palette=pal, ax=ax)
        ax.set_title(title)
fig.tight_layout()
fig.savefig('palettes.png', dpi=100, bbox_inches='tight')
print('Saved palettes.png')
plt.close()

import seaborn as sns
import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

np.random.seed(42)
df = pd.DataFrame({'x':np.repeat(range(5),30),'y':np.random.randn(150)+np.repeat([1,2,3,4,5],30)})
fig, axes = plt.subplots(1, 2, figsize=(12,5))
with sns.axes_style('ticks'):
    sns.boxplot(data=df, x='x', y='y', ax=axes[0], palette='pastel')
    sns.despine(ax=axes[0], trim=True)
    axes[0].set_title('Despined + trimmed')
with sns.axes_style('whitegrid'):
    sns.violinplot(data=df, x='x', y='y', ax=axes[1], palette='muted', inner='box')
    sns.despine(ax=axes[1], left=False, bottom=False, top=True, right=True)
    axes[1].set_title('Whitegrid + partial despine')
fig.tight_layout()
fig.savefig('despine_styles.png', dpi=100, bbox_inches='tight')
print('Saved despine_styles.png')
plt.close()

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(11, 5))
sns.stripplot(data=tips, x='day', y='total_bill', ax=ax1,
              palette='Set2', jitter=True, alpha=0.7, size=5)
ax1.set_title('Strip Plot (jitter)')

sns.swarmplot(data=tips, x='day', y='total_bill', ax=ax2,
              palette='Set2', size=4)
ax2.set_title('Swarm Plot (no overlap)')

for ax in (ax1, ax2):
    ax.set_xlabel('Day'); ax.set_ylabel('Total Bill ($)')
fig.tight_layout()
fig.savefig('strip_swarm.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved strip_swarm.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

fig, ax = plt.subplots(figsize=(8, 5))
sns.boxplot(data=iris, x='species', y='sepal_length', ax=ax,
            palette='pastel', width=0.5,
            boxprops=dict(alpha=0.6))
sns.stripplot(data=iris, x='species', y='sepal_length', ax=ax,
              palette='Set2', size=5, jitter=True, alpha=0.7,
              linewidth=0.5, edgecolor='gray')
ax.set_title('Box + Strip Plot Overlay')
ax.set_xlabel('Species'); ax.set_ylabel('Sepal Length (cm)')
fig.tight_layout()
fig.savefig('box_strip.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved box_strip.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(0)
df = pd.DataFrame({
    'score': np.concatenate([np.random.normal(m, 1, 60) for m in [5, 6.5, 8]]),
    'group': np.repeat(['Control', 'Low Dose', 'High Dose'], 60)
})

fig, ax = plt.subplots(figsize=(8, 5))
sns.violinplot(data=df, x='group', y='score', ax=ax,
               palette='muted', inner=None, alpha=0.6)
sns.swarmplot(data=df, x='group', y='score', ax=ax,
              color='black', size=3, alpha=0.6)
ax.set_title('Violin + Swarm: Treatment Groups')
ax.set_xlabel('Group'); ax.set_ylabel('Score')
fig.tight_layout()
fig.savefig('violin_swarm.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved violin_swarm.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, ax = plt.subplots(figsize=(9, 5))
sns.stripplot(data=tips, x='day', y='total_bill', hue='sex',
              dodge=True, jitter=True, alpha=0.7, size=5,
              palette='Set1', ax=ax)
ax.set_title('Strip Plot: Tip Amount by Day and Sex (Dodged)')
ax.set_xlabel('Day'); ax.set_ylabel('Total Bill ($)')
ax.legend(title='Sex', bbox_to_anchor=(1, 1))
fig.tight_layout()
fig.savefig('strip_hue.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved strip_hue.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(11, 5))
sns.pointplot(data=tips, x='day', y='total_bill', ax=ax1,
              palette='Set2', capsize=0.1, errwidth=2,
              markers='o', linestyles='-')
ax1.set_title('Point Plot: Mean Bill by Day')

sns.pointplot(data=tips, x='day', y='total_bill', hue='sex',
              ax=ax2, palette='Set1', dodge=True,
              capsize=0.08, errwidth=1.5)
ax2.set_title('Point Plot: By Day and Sex')

for ax in (ax1, ax2):
    ax.set_xlabel('Day'); ax.set_ylabel('Mean Total Bill ($)')
fig.tight_layout()
fig.savefig('pointplot.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved pointplot.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='darkgrid')
np.random.seed(0)
n = 50
df = pd.DataFrame({
    't': np.tile(np.arange(n), 3),
    'value': np.concatenate([
        np.cumsum(np.random.randn(n)) + 5,
        np.cumsum(np.random.randn(n)) + 3,
        np.cumsum(np.random.randn(n)) + 7,
    ]),
    'model': np.repeat(['Model A', 'Model B', 'Model C'], n)
})

# Add repeated measurements for CI
df_rep = pd.concat([df.assign(value=df.value + np.random.randn(len(df))*0.5)
                    for _ in range(5)], ignore_index=True)

fig, ax = plt.subplots(figsize=(10, 5))
sns.lineplot(data=df_rep, x='t', y='value', hue='model',
             palette='Set2', linewidth=2, ax=ax)
ax.set_title('lineplot with 95% CI Shading')
ax.set_xlabel('Time Step'); ax.set_ylabel('Value')
fig.tight_layout()
fig.savefig('lineplot_ci.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved lineplot_ci.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(1)
months = list(range(1, 13))
df = pd.DataFrame({
    'month': months * 3,
    'revenue': (
        [100 + i*8 + np.random.randn()*5 for i in range(12)] +
        [80 + i*6 + np.random.randn()*4 for i in range(12)] +
        [60 + i*10 + np.random.randn()*6 for i in range(12)]
    ),
    'region': ['North']*12 + ['South']*12 + ['East']*12
})

fig, ax = plt.subplots(figsize=(10, 5))
sns.lineplot(data=df, x='month', y='revenue', hue='region',
             style='region', markers=True, dashes=False,
             palette='Set1', linewidth=2, ax=ax)
ax.set_title('Monthly Revenue by Region')
ax.set_xlabel('Month'); ax.set_ylabel('Revenue ($K)')
ax.legend(title='Region')
fig.tight_layout()
fig.savefig('lineplot_markers.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved lineplot_markers.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
titanic = sns.load_dataset('titanic')

fig, axes = plt.subplots(1, 3, figsize=(14, 5))
# Bar: shows aggregated height
sns.barplot(data=titanic, x='class', y='survived', hue='sex',
            palette='Set2', ax=axes[0])
axes[0].set_title('barplot: Mean Survival Rate')

# Point: cleaner for comparison
sns.pointplot(data=titanic, x='class', y='survived', hue='sex',
              palette='Set1', dodge=True, capsize=0.1, ax=axes[1])
axes[1].set_title('pointplot: Same Data')

# Count the actual survivors
surv = titanic.groupby(['class','sex'])['survived'].mean().reset_index()
sns.pointplot(data=surv, x='class', y='survived', hue='sex',
              palette='Set1', dodge=True, capsize=0.1, ax=axes[2],
              markers=['o', 's'], linestyles=['-','--'])
axes[2].set_title('pointplot: Precomputed')

for ax in axes:
    ax.set_ylim(0, 1); ax.set_ylabel('Survival Rate')
fig.tight_layout()
fig.savefig('point_vs_bar.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved point_vs_bar.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(0)
df = pd.DataFrame({
    'value': np.concatenate([
        np.random.normal(5, 1.5, 200),
        np.random.normal(7, 1.0, 200),
        np.random.exponential(2, 200) + 3,
    ]),
    'group': np.repeat(['Normal(5,1.5)', 'Normal(7,1)', 'Exp+3'], 200)
})

fig, ax = plt.subplots(figsize=(9, 5))
sns.ecdfplot(data=df, x='value', hue='group', palette='Set2', linewidth=2)
ax.axhline(0.5, color='gray', linestyle='--', linewidth=1, label='Median')
ax.set_title('ECDF: Distribution Comparison')
ax.set_xlabel('Value'); ax.set_ylabel('Cumulative Proportion')
ax.legend(title='Group')
fig.tight_layout()
fig.savefig('ecdf.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved ecdf.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

fig, axes = plt.subplots(1, 2, figsize=(12, 5))
# Separate feature distributions per species
for feat, ax in zip(['sepal_length', 'petal_length'], axes):
    sns.kdeplot(data=iris, x=feat, hue='species',
                fill=True, alpha=0.35, linewidth=2, palette='Set2', ax=ax)
    ax.set_title(f'KDE: {feat.replace("_"," ").title()}')
    ax.set_xlabel(feat.replace('_',' ').title())
fig.tight_layout()
fig.savefig('kde_compare.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved kde_compare.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(1)
df = pd.DataFrame({
    'response_ms': np.concatenate([
        np.random.lognormal(5, 0.5, 300),
        np.random.lognormal(5.5, 0.4, 200),
    ]),
    'endpoint': np.repeat(['/api/search', '/api/checkout'], [300, 200])
})

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))
sns.histplot(data=df, x='response_ms', hue='endpoint',
             stat='density', kde=True, alpha=0.4, palette='Set1',
             common_norm=False, ax=ax1)
ax1.set_title('Histogram with KDE — Linear scale')
ax1.set_xlabel('Response Time (ms)')

sns.histplot(data=df, x='response_ms', hue='endpoint',
             stat='density', kde=True, alpha=0.4, palette='Set1',
             common_norm=False, log_scale=True, ax=ax2)
ax2.set_title('Histogram with KDE — Log scale')
ax2.set_xlabel('Response Time (ms)')
fig.tight_layout()
fig.savefig('hist_kde_compare.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved hist_kde_compare.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(2)
data = pd.DataFrame({
    'latency_ms': np.random.lognormal(5, 0.7, 1000),
    'server': np.random.choice(['US-East', 'EU-West', 'AP-South'], 1000)
})

fig, ax = plt.subplots(figsize=(9, 5))
sns.ecdfplot(data=data, x='latency_ms', hue='server',
             palette='tab10', linewidth=2, ax=ax)

# Mark p50, p95, p99
for pct, label, color in [(50,'p50','gray'),(95,'p95','orange'),(99,'p99','red')]:
    val = np.percentile(data['latency_ms'], pct)
    ax.axvline(val, color=color, linestyle='--', linewidth=1.2, alpha=0.8)
    ax.axhline(pct/100, color=color, linestyle=':', linewidth=0.8, alpha=0.5)
    ax.text(val+50, pct/100-0.04, f'{label}: {val:.0f}ms',
            fontsize=8, color=color)

ax.set_title('API Latency ECDF by Server Region')
ax.set_xlabel('Latency (ms)'); ax.set_ylabel('Cumulative Proportion')
fig.tight_layout()
fig.savefig('ecdf_percentiles.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved ecdf_percentiles.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
iris = sns.load_dataset('iris')
corr = iris.drop('species', axis=1).corr()

g = sns.clustermap(corr, cmap='RdBu_r', vmin=-1, vmax=1,
                   annot=True, fmt='.2f', figsize=(7, 7),
                   linewidths=0.5)
g.ax_heatmap.set_title('Iris Feature Correlation Clustermap', pad=50)
plt.savefig('clustermap_iris.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved clustermap_iris.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(0)
n_genes, n_samples = 20, 12
groups = ['A','A','A','A','B','B','B','B','C','C','C','C']
data = pd.DataFrame(
    np.random.randn(n_genes, n_samples) +
    np.array([np.sin(np.linspace(0,3,n_samples))*(i%3-1) for i in range(n_genes)]),
    index=[f'Gene_{i:02d}' for i in range(n_genes)],
    columns=[f'S{i+1:02d}' for i in range(n_samples)]
)
palette = {'A':'#4c72b0', 'B':'#dd8452', 'C':'#55a868'}
col_colors = pd.Series(groups, index=data.columns).map(palette)

g = sns.clustermap(data, cmap='vlag', figsize=(10, 8),
                   col_colors=col_colors,
                   row_cluster=True, col_cluster=True,
                   z_score=0, linewidths=0.3)
g.ax_heatmap.set_title('Gene Expression Clustermap', pad=50)
plt.savefig('clustermap_genes.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved clustermap_genes.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(1)
features = ['Revenue', 'Growth', 'Margin', 'CAC', 'LTV', 'Churn']
companies = [f'Co{i:02d}' for i in range(1, 16)]
data = pd.DataFrame(
    np.random.randn(len(companies), len(features)),
    index=companies, columns=features
)
# Add cluster structure
data.iloc[:5, :3] += 2
data.iloc[5:10, 3:] -= 2
data.iloc[10:, [0,2,4]] += 1.5

g = sns.clustermap(data, cmap='coolwarm', figsize=(8, 10),
                   standard_scale=1,  # standardize each column
                   linewidths=0.4, annot=False,
                   dendrogram_ratio=(0.15, 0.2))
g.ax_heatmap.set_title('Company KPI Clustermap
(column-standardized)', pad=50)
plt.savefig('clustermap_kpi.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved clustermap_kpi.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(2)
data = pd.DataFrame(
    np.random.randn(15, 10),
    index=[f'R{i}' for i in range(15)],
    columns=[f'C{j}' for j in range(10)]
)
data.iloc[:5, :4] += 2.5
data.iloc[10:, 6:] += 2.5

fig, axes_list = plt.subplots(1, 2, figsize=(14, 6))
for ax, (method, metric) in zip(axes_list, [('ward','euclidean'),('average','correlation')]):
    g = sns.clustermap(data, cmap='RdYlBu_r', figsize=(6, 5),
                       method=method, metric=metric,
                       linewidths=0.3)
    g.fig.suptitle(f'method={method}, metric={metric}', y=1.01, fontsize=10)
    g.fig.savefig(f'clustermap_{method}.png', dpi=100, bbox_inches='tight')
    plt.close(g.fig)
print('Saved clustermap_ward.png and clustermap_average.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

kinds = ['scatter', 'hex', 'kde', 'reg']
for kind in kinds:
    g = sns.jointplot(data=tips, x='total_bill', y='tip',
                      kind=kind, palette='Set2', height=5,
                      marginal_kws=dict(bins=25))
    g.fig.suptitle(f"kind='{kind}'", y=1.02)
    g.fig.savefig(f'joint_{kind}.png', dpi=100, bbox_inches='tight')
    plt.close(g.fig)
print('Saved joint_scatter/hex/kde/reg.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

g = sns.jointplot(data=iris, x='sepal_length', y='petal_length',
                  hue='species', palette='Set2',
                  height=6, marginal_kws=dict(fill=True, alpha=0.4))
g.fig.suptitle('Iris: Sepal vs Petal Length (hue=species)', y=1.02)
g.fig.savefig('joint_hue.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved joint_hue.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(42)
df = pd.DataFrame({
    'x': np.random.lognormal(2, 0.5, 300),
    'y': np.random.lognormal(1.5, 0.6, 300)
})

g = sns.JointGrid(data=df, x='x', y='y', height=6)
g.plot_joint(sns.scatterplot, alpha=0.4, color='steelblue', s=25)
g.plot_joint(sns.kdeplot, levels=5, color='navy', linewidths=1.0)
g.plot_marginals(sns.histplot, kde=True, bins=25, color='steelblue', alpha=0.6)

g.ax_joint.set_xlabel('X (lognormal)'); g.ax_joint.set_ylabel('Y (lognormal)')
g.fig.suptitle('JointGrid: Scatter + KDE + Histogram Marginals', y=1.01)
g.fig.savefig('joint_grid_custom.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved joint_grid_custom.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(5)
n = 150
x = np.random.uniform(10, 100, n)
y = 0.5*x + np.random.randn(n)*8 + 5

df = pd.DataFrame({'x': x, 'y': y})

g = sns.jointplot(data=df, x='x', y='y',
                  kind='reg', height=6,
                  scatter_kws=dict(alpha=0.5, s=30, color='steelblue'),
                  line_kws=dict(color='red', linewidth=2),
                  marginal_kws=dict(bins=20, kde=True))
# Compute and annotate correlation
r = df.corr().iloc[0,1]
g.ax_joint.text(0.05, 0.92, f'r = {r:.3f}', transform=g.ax_joint.transAxes,
                fontsize=11, fontweight='bold', color='red')
g.fig.suptitle('Joint Regression Plot with Correlation', y=1.02)
g.fig.savefig('joint_reg.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved joint_reg.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

g = sns.catplot(data=tips, x='day', y='total_bill',
                hue='sex', col='time',
                kind='box', palette='Set2',
                height=5, aspect=0.85,
                order=['Thur','Fri','Sat','Sun'])
g.set_axis_labels('Day', 'Total Bill ($)')
g.set_titles('{col_name}')
g.fig.suptitle('Bill by Day, Sex, and Time (catplot col)', y=1.02, fontweight='bold')
g.fig.savefig('catplot_col.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved catplot_col.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
titanic = sns.load_dataset('titanic')

g = sns.catplot(data=titanic, x='class', y='age',
                col='survived', hue='sex',
                kind='violin', split=True, inner='quartile',
                palette='pastel', height=5, aspect=0.9)
g.set_axis_labels('Class', 'Age')
g.set_titles(col_template='Survived: {col_name}')
g.fig.suptitle('Titanic Age by Class, Gender & Survival', y=1.02, fontweight='bold')
g.fig.savefig('catplot_violin.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved catplot_violin.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

fig, axes = plt.subplots(1, 2, figsize=(13, 5))
# Count plot via catplot saved to file; replicate on axes
sns.countplot(data=tips, x='day', hue='sex', palette='Set1', ax=axes[0])
axes[0].set_title("countplot: Visits by Day")

sns.barplot(data=tips, x='day', y='tip', hue='sex',
            palette='Set2', capsize=0.08, ax=axes[1])
axes[1].set_title("barplot: Mean Tip by Day")
for ax in axes:
    ax.set_xlabel('Day')
fig.tight_layout()
fig.savefig('catplot_count_bar.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved catplot_count_bar.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
titanic = sns.load_dataset('titanic')

g = sns.catplot(data=titanic, x='class', y='survived',
                hue='sex', col='embarked',
                kind='point', dodge=True, capsize=0.1,
                markers=['o','s'], linestyles=['-','--'],
                palette='Set1', height=4, aspect=0.85,
                order=['First','Second','Third'])
g.set_axis_labels('Class', 'Survival Rate')
g.set_titles(col_template='Embarked: {col_name}')
g.fig.suptitle('Survival Rate by Class, Sex & Port', y=1.02, fontweight='bold')
g.fig.savefig('catplot_point.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved catplot_point.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(0)
df = pd.DataFrame({
    'value': np.concatenate([np.random.normal(0,1,300), np.random.normal(4,1.5,300)]),
    'group': np.repeat(['A','B'], 300)
})

for kind in ['hist','kde','ecdf']:
    g = sns.displot(data=df, x='value', hue='group',
                    kind=kind, height=4, aspect=1.5, palette='Set2',
                    fill=(kind != 'ecdf'), alpha=0.5, linewidth=2)
    g.fig.suptitle(f"displot kind='{kind}'", y=1.02)
    g.fig.savefig(f'displot_{kind}.png', dpi=100, bbox_inches='tight')
    plt.close(g.fig)
print('Saved displot_hist/kde/ecdf.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
titanic = sns.load_dataset('titanic').dropna(subset=['age'])

g = sns.displot(data=titanic, x='age', col='class',
                hue='survived', kind='hist', stat='density',
                kde=True, alpha=0.5, palette='Set1',
                height=4, aspect=0.85, col_order=['First','Second','Third'])
g.set_axis_labels('Age', 'Density')
g.set_titles(col_template='{col_name} Class')
g.fig.suptitle('Age Distribution by Class and Survival', y=1.02, fontweight='bold')
g.fig.savefig('displot_col.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved displot_col.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(1)
regions = ['North','South']
periods = ['Q1','Q2','Q3','Q4']
rows = []
for region in regions:
    base = 50 if region == 'North' else 35
    for period in periods:
        offset = periods.index(period) * 5
        vals = np.random.normal(base + offset, 12, 100)
        for v in vals:
            rows.append({'region': region, 'period': period, 'sales': v})
df = pd.DataFrame(rows)

g = sns.displot(data=df, x='sales', row='region', col='period',
                kind='kde', fill=True, alpha=0.5, palette='Set2',
                height=3, aspect=1.1, col_order=periods)
g.set_axis_labels('Sales ($K)', 'Density')
g.fig.suptitle('Sales Distribution by Region & Quarter', y=1.02, fontweight='bold')
g.fig.savefig('displot_grid.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved displot_grid.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(2)
df = pd.DataFrame({
    'latency': np.concatenate([
        np.random.lognormal(3, 0.4, 500),
        np.random.lognormal(4, 0.5, 200),
    ]),
    'endpoint': np.repeat(['/fast','/slow'], [500,200])
})

g = sns.displot(data=df, x='latency', hue='endpoint',
                kind='hist', stat='density', kde=True,
                rug=True, rug_kws=dict(height=0.05, alpha=0.3),
                log_scale=True, alpha=0.4, palette='Set1',
                height=5, aspect=1.6, binwidth=0.05)
g.set_axis_labels('Latency (ms, log scale)', 'Density')
g.fig.suptitle('Endpoint Latency Distribution with Rug', y=1.02, fontweight='bold')
g.fig.savefig('displot_rug.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved displot_rug.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

g = sns.relplot(data=tips, x='total_bill', y='tip',
                hue='sex', col='time', style='sex',
                palette='Set1', s=60, alpha=0.7,
                height=4, aspect=1.0)
g.set_axis_labels('Total Bill ($)', 'Tip ($)')
g.set_titles(col_template='{col_name}')
g.fig.suptitle('Tips: Scatter by Time and Sex (relplot)', y=1.02, fontweight='bold')
g.fig.savefig('relplot_scatter.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved relplot_scatter.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='darkgrid')
np.random.seed(0)
subjects = [f'Subject_{i:02d}' for i in range(1, 7)]
df_list = []
for subj in subjects:
    t = np.arange(20)
    y = np.sin(t*0.3 + np.random.uniform(0,3)) + np.random.randn(20)*0.2
    df_list.append(pd.DataFrame({'time':t,'value':y,'subject':subj}))
df = pd.concat(df_list, ignore_index=True)

g = sns.relplot(data=df, x='time', y='value', col='subject',
                kind='line', col_wrap=3, palette='tab10',
                height=3, aspect=1.3, linewidth=2)
g.set_axis_labels('Time', 'Signal')
g.set_titles(col_template='{col_name}')
g.fig.suptitle('Subject Time Series (relplot col_wrap=3)', y=1.02, fontweight='bold')
g.fig.savefig('relplot_line.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved relplot_line.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(1)
n = 200
df = pd.DataFrame({
    'x': np.random.randn(n),
    'y': np.random.randn(n),
    'size_var': np.random.uniform(50, 300, n),
    'category': np.random.choice(['A','B','C'], n),
    'quality': np.random.choice(['High','Low'], n)
})

g = sns.relplot(data=df, x='x', y='y',
                hue='category', size='size_var', style='quality',
                palette='Set2', sizes=(20, 300), alpha=0.7,
                height=5, aspect=1.2)
g.ax.set_title('Multi-Encoding Scatter: hue + size + style', fontweight='bold')
g.fig.savefig('relplot_multi.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved relplot_multi.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(2)
conditions = ['Fast','Slow']
groups = ['Treatment','Control']
rows = []
for cond in conditions:
    for grp in groups:
        mu = (4 if grp=='Treatment' else 2) + (1 if cond=='Fast' else 0)
        t = np.arange(30)
        y = mu + np.cumsum(np.random.randn(30)*0.3) + np.sin(t*0.2)
        df_part = pd.DataFrame({'time':t,'signal':y,
                                 'condition':cond,'group':grp})
        rows.append(df_part)
df = pd.concat(rows, ignore_index=True)

g = sns.relplot(data=df, x='time', y='signal',
                row='group', col='condition',
                kind='line', palette='Set1', hue='group',
                height=3.5, aspect=1.2, linewidth=2)
g.set_axis_labels('Time', 'Signal')
g.set_titles(row_template='{row_name}', col_template='{col_name}')
g.fig.suptitle('Signal by Group × Condition (relplot grid)', y=1.02, fontweight='bold')
g.fig.savefig('relplot_grid.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved relplot_grid.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(0)
n = 8
labels = [f'Var_{i}' for i in range(1, n+1)]
data = np.random.randn(200, n)
for i in range(n):
    for j in range(i):
        data[:, j] += data[:, i] * np.random.uniform(-0.5, 0.8)
corr = np.corrcoef(data.T)
df_corr = pd.DataFrame(corr, index=labels, columns=labels)

mask = np.triu(np.ones_like(corr, dtype=bool), k=1)  # mask upper

fig, ax = plt.subplots(figsize=(8, 7))
sns.heatmap(df_corr, mask=mask, cmap='RdBu_r', vmin=-1, vmax=1,
            annot=True, fmt='.2f', linewidths=0.5,
            ax=ax, square=True, cbar_kws={'label': 'Pearson r'})
ax.set_title('Lower-Triangle Correlation Matrix', fontweight='bold')
fig.tight_layout()
fig.savefig('corr_heatmap.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved corr_heatmap.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(1)
days = ['Mon','Tue','Wed','Thu','Fri','Sat','Sun']
hours = [f'{h:02d}:00' for h in range(7, 23)]
traffic = np.random.poisson(50, (len(hours), len(days)))
traffic[:, 5:] = np.random.poisson(80, (len(hours), 2))
traffic[8:12, :] += 30
traffic[17:20, :] += 40

df = pd.DataFrame(traffic, index=hours, columns=days)
fig, ax = plt.subplots(figsize=(10, 7))
sns.heatmap(df, cmap='YlOrRd', annot=True, fmt='d',
            linewidths=0.3, ax=ax,
            cbar_kws={'label': 'Page Views', 'shrink': 0.8})
ax.set_title('Website Traffic by Hour & Day', fontweight='bold')
ax.set_xlabel('Day of Week'); ax.set_ylabel('Hour')
fig.tight_layout()
fig.savefig('traffic_heatmap.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved traffic_heatmap.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(2)
models = ['LR','RF','XGB','SVM','NN']
metrics = ['Accuracy','Precision','Recall','F1','AUC']
values = np.random.uniform(0.70, 0.98, (len(models), len(metrics)))
df = pd.DataFrame(values, index=models, columns=metrics)

# Custom annotations: highlight best per metric
annot = df.copy().applymap(lambda v: f'{v:.3f}')
best_rows = df.idxmax(axis=0)
highlights = pd.DataFrame('', index=df.index, columns=df.columns)
for metric in metrics:
    highlights.loc[best_rows[metric], metric] = f'★{df.loc[best_rows[metric],metric]:.3f}'

fig, ax = plt.subplots(figsize=(9, 5))
sns.heatmap(df, cmap='Blues', vmin=0.65, vmax=1.0, ax=ax,
            annot=df.round(3), fmt='', linewidths=0.5)
ax.set_title('Model Comparison Heatmap (★ = best per metric)', fontweight='bold')
fig.tight_layout()
fig.savefig('model_heatmap.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved model_heatmap.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(3)
products = [f'P{i:02d}' for i in range(1, 9)]
kpis = ['Revenue','Margin','Units','Traffic','Conv%','Retention']
changes = np.random.randn(len(products), len(kpis)) * 15
df = pd.DataFrame(changes, index=products, columns=kpis)

fig, ax = plt.subplots(figsize=(9, 6))
sns.heatmap(df, cmap='RdYlGn', center=0,
            annot=True, fmt='.1f', linewidths=0.5, ax=ax,
            cbar_kws={'label': 'YoY Change (%)'})
ax.set_title('Product KPI Year-over-Year Change (%)', fontweight='bold')
fig.tight_layout()
fig.savefig('diverging_heatmap.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved diverging_heatmap.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

g = sns.PairGrid(iris, hue='species', palette='Set2',
                 vars=['sepal_length','sepal_width','petal_length','petal_width'])
g.map_diag(sns.histplot, kde=True, alpha=0.6)
g.map_offdiag(sns.scatterplot, s=25, alpha=0.6)
g.add_legend(title='Species')
g.fig.suptitle('PairGrid: Histogram Diagonal + Scatter', y=1.01, fontweight='bold')
g.fig.savefig('pairgrid_hist.png', dpi=100, bbox_inches='tight')
plt.close(g.fig)
print('Saved pairgrid_hist.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')
vars_ = ['sepal_length','sepal_width','petal_length','petal_width']

g = sns.PairGrid(iris, hue='species', palette='Set1', vars=vars_)
g.map_diag(sns.kdeplot, fill=True, alpha=0.4, linewidth=1.5)
g.map_upper(sns.kdeplot, levels=4, warn_singular=False)
g.map_lower(sns.scatterplot, s=20, alpha=0.5)
g.add_legend(title='Species', fontsize=9)
g.fig.suptitle('PairGrid: KDE + Scatter Split', y=1.01, fontweight='bold')
g.fig.savefig('pairgrid_split.png', dpi=100, bbox_inches='tight')
plt.close(g.fig)
print('Saved pairgrid_split.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')
vars_ = ['sepal_length','petal_length','petal_width']

g = sns.PairGrid(iris, hue='species', palette='Set2', vars=vars_)

def diag_box(x, **kwargs):
    ax = plt.gca()
    data = kwargs.get('data', x)
    groups = x.groupby(level=0) if hasattr(x,'groupby') else None
    ax.boxplot([x[x.index == i] for i in x.unique()], vert=True)

g.map_diag(sns.kdeplot, fill=True, alpha=0.5)
g.map_upper(sns.scatterplot, s=20, alpha=0.5)
g.map_lower(sns.regplot, scatter_kws=dict(s=10,alpha=0.4),
            line_kws=dict(linewidth=1.5))
g.add_legend()
g.fig.suptitle('PairGrid: KDE / Scatter / Regression', y=1.01, fontweight='bold')
g.fig.savefig('pairgrid_reg.png', dpi=100, bbox_inches='tight')
plt.close(g.fig)
print('Saved pairgrid_reg.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(0)
n = 150
df = pd.DataFrame({
    'A': np.random.randn(n),
    'B': np.random.randn(n),
    'C': np.random.randn(n),
    'group': np.random.choice(['X','Y'], n)
})
df['B'] += df['A'] * 0.7
df['C'] = df['A'] * (-0.5) + np.random.randn(n) * 0.7

def corrfunc(x, y, **kwargs):
    r = np.corrcoef(x, y)[0,1]
    ax = plt.gca()
    ax.annotate(f'r = {r:.2f}', xy=(0.5,0.5), xycoords='axes fraction',
                ha='center', va='center', fontsize=12,
                color='darkred' if abs(r) > 0.4 else 'gray',
                fontweight='bold' if abs(r) > 0.4 else 'normal')

g = sns.PairGrid(df[['A','B','C','group']], hue='group', palette='Set1',
                 vars=['A','B','C'])
g.map_diag(sns.kdeplot, fill=True, alpha=0.4)
g.map_lower(sns.scatterplot, s=20, alpha=0.5)
g.map_upper(corrfunc)
g.add_legend()
g.fig.suptitle('PairGrid: Lower=Scatter, Upper=Correlation', y=1.01, fontweight='bold')
g.fig.savefig('pairgrid_corr.png', dpi=100, bbox_inches='tight')
plt.close(g.fig)
print('Saved pairgrid_corr.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(0)
n = 150
x = np.linspace(0, 10, n)
y_linear  = 2*x + 1 + np.random.randn(n)*2
y_nonlin  = 2*x + 0.3*x**2 + np.random.randn(n)*3

df = pd.DataFrame({'x':x,'y_linear':y_linear,'y_nonlin':y_nonlin})

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 5))
sns.residplot(data=df, x='x', y='y_linear', lowess=True,
              scatter_kws=dict(alpha=0.5, s=20),
              line_kws=dict(color='red', linewidth=2), ax=ax1)
ax1.set_title('Residuals: Linear Data (good fit)')
ax1.axhline(0, color='gray', linestyle='--', linewidth=1)

sns.residplot(data=df, x='x', y='y_nonlin', lowess=True,
              scatter_kws=dict(alpha=0.5, s=20),
              line_kws=dict(color='red', linewidth=2), ax=ax2)
ax2.set_title('Residuals: Nonlinear Data (pattern visible)')
ax2.axhline(0, color='gray', linestyle='--', linewidth=1)

fig.suptitle('residplot — Lowess Smoother', fontweight='bold')
fig.tight_layout()
fig.savefig('residplot.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved residplot.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

g = sns.lmplot(data=tips, x='total_bill', y='tip',
               hue='sex', palette='Set1',
               scatter_kws=dict(s=30, alpha=0.6),
               line_kws=dict(linewidth=2),
               height=5, aspect=1.3)
g.set_axis_labels('Total Bill ($)', 'Tip ($)')
g.fig.suptitle('lmplot: Regression by Sex', y=1.02, fontweight='bold')
g.fig.savefig('lmplot_hue.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved lmplot_hue.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

g = sns.lmplot(data=tips, x='total_bill', y='tip',
               col='time', hue='smoker',
               palette='Set2',
               scatter_kws=dict(s=30, alpha=0.6),
               height=4, aspect=1.1)
g.set_axis_labels('Total Bill ($)', 'Tip ($)')
g.set_titles(col_template='{col_name}')
g.fig.suptitle('lmplot: Regression by Time and Smoker', y=1.02, fontweight='bold')
g.fig.savefig('lmplot_col.png', dpi=120, bbox_inches='tight')
plt.close(g.fig)
print('Saved lmplot_col.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(1)
x = np.linspace(-3, 3, 100)
y = 2*x**3 - x**2 + x + np.random.randn(100)*3
df = pd.DataFrame({'x':x,'y':y})

fig, axes = plt.subplots(1, 3, figsize=(13, 4))
for ax, order, title in zip(axes, [1,2,3], ['Linear (order=1)','Quadratic (order=2)','Cubic (order=3)']):
    sns.regplot(data=df, x='x', y='y', order=order,
                scatter_kws=dict(s=15, alpha=0.5),
                line_kws=dict(linewidth=2, color='tomato'),
                ax=ax)
    ax.set_title(title)
    ax.grid(True, alpha=0.3)
fig.suptitle('Polynomial Regression Orders with regplot', fontweight='bold')
fig.tight_layout()
fig.savefig('poly_regplot.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved poly_regplot.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(0)
df = pd.DataFrame({
    'score': np.random.normal(72, 15, 200),
    'group': np.random.choice(['A','B','C','D'], 200)
})

fig, ax = plt.subplots(figsize=(9, 5))
sns.boxplot(data=df, x='group', y='score', palette='Set2', ax=ax)

# Matplotlib overlays
ax.axhline(70, color='red', linestyle='--', linewidth=2, label='Target')
ax.axhspan(0, 60, color='red', alpha=0.07, label='Fail zone')
ax.axhspan(90, 100, color='green', alpha=0.07, label='Excellent')
ax.set_title('Test Scores by Group with Reference Lines', fontweight='bold')
ax.legend()
ax.set_ylim(20, 110)
fig.tight_layout()
fig.savefig('sns_mpl_lines.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved sns_mpl_lines.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(1)
df = pd.DataFrame({
    'revenue': np.concatenate([np.random.normal(100, 20, 95), [220, 230, 240, 10, 5]]),
    'region': np.random.choice(['East','West'], 100)
})

fig, ax = plt.subplots(figsize=(9, 5))
sns.stripplot(data=df, x='region', y='revenue', jitter=True,
              palette='Set2', size=6, alpha=0.7, ax=ax)

# Annotate extreme outliers
q1, q3 = df.revenue.quantile([0.25, 0.75])
iqr = q3 - q1
outliers = df[df.revenue > q3 + 1.5*iqr]
for _, row in outliers.iterrows():
    x_jit = {'East':0,'West':1}[row.region] + np.random.uniform(-0.1,0.1)
    ax.annotate(f'${row.revenue:.0f}K',
                xy=(x_jit, row.revenue),
                xytext=(x_jit+0.2, row.revenue),
                fontsize=8, color='darkred',
                arrowprops=dict(arrowstyle='->', color='darkred', lw=1))

ax.axhline(q3+1.5*iqr, color='red', linestyle='--', linewidth=1.2, label='Upper fence')
ax.set_title('Revenue Distribution with Outlier Annotations', fontweight='bold')
ax.legend()
fig.tight_layout()
fig.savefig('sns_annotate_outliers.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved sns_annotate_outliers.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

import matplotlib.patches as mpatches

sns.set_theme(style='white')
np.random.seed(2)
data = np.random.randn(8, 8)
data[2:4, 5:7] += 3  # hot cluster

fig, ax = plt.subplots(figsize=(7, 6))
sns.heatmap(data, cmap='RdBu_r', center=0, ax=ax,
            annot=True, fmt='.1f', linewidths=0.3)

# Highlight cluster with rectangle
rect = mpatches.Rectangle((5, 2), 2, 2,
    linewidth=3, edgecolor='gold', facecolor='none',
    transform=ax.transData)
ax.add_patch(rect)
ax.text(6.5, 1.7, 'Hot cluster', ha='center', fontsize=10,
        color='gold', fontweight='bold')
ax.set_title('Heatmap with Cluster Highlight', fontweight='bold')
fig.tight_layout()
fig.savefig('sns_heatmap_patch.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved sns_heatmap_patch.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(3)
months = list(range(1, 13))
df = pd.DataFrame({
    'month': months,
    'revenue': [80+i*5+np.random.randn()*4 for i in range(12)],
    'users': [1000+i*80+np.random.randn()*50 for i in range(12)]
})

fig, ax1 = plt.subplots(figsize=(10, 5))
ax2 = ax1.twinx()

sns.lineplot(data=df, x='month', y='revenue', color='steelblue',
             linewidth=2, marker='o', label='Revenue ($K)', ax=ax1)
sns.lineplot(data=df, x='month', y='users', color='tomato',
             linewidth=2, marker='s', linestyle='--', label='Users', ax=ax2)

ax1.set_ylabel('Revenue ($K)', color='steelblue')
ax2.set_ylabel('Users', color='tomato')
ax1.tick_params(axis='y', labelcolor='steelblue')
ax2.tick_params(axis='y', labelcolor='tomato')
ax1.set_xlabel('Month')

lines1, lab1 = ax1.get_legend_handles_labels()
lines2, lab2 = ax2.get_legend_handles_labels()
ax1.legend(lines1+lines2, lab1+lab2, loc='upper left')
ax1.get_legend().remove() if ax2.get_legend() else None
ax1.set_title('Revenue & Users (seaborn + twinx)', fontweight='bold')
fig.tight_layout()
fig.savefig('sns_twinx.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved sns_twinx.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
titanic = sns.load_dataset('titanic')

fig, axes = plt.subplots(2, 2, figsize=(12, 9))
# Survival by class
sns.barplot(data=titanic, x='class', y='survived', hue='sex',
            palette='Set1', capsize=0.08, ax=axes[0,0])
axes[0,0].set_title('Survival Rate by Class & Sex')
axes[0,0].set_ylim(0,1)

# Age distribution by survival
sns.kdeplot(data=titanic.dropna(subset=['age']), x='age', hue='survived',
            fill=True, alpha=0.4, palette='Set2', ax=axes[0,1])
axes[0,1].set_title('Age Distribution by Survival')

# Fare vs survival (box)
sns.boxplot(data=titanic, x='class', y='fare', hue='survived',
            palette='pastel', ax=axes[1,0])
axes[1,0].set_title('Fare Distribution by Class & Survival')

# Count by embarkation port
sns.countplot(data=titanic, x='embarked', hue='survived',
              palette='Set2', ax=axes[1,1])
axes[1,1].set_title('Counts by Port of Embarkation')

fig.suptitle('Titanic Dataset — EDA Dashboard', fontweight='bold', fontsize=13)
fig.tight_layout()
fig.savefig('titanic_eda.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved titanic_eda.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
diamonds = sns.load_dataset('diamonds')
sample = diamonds.sample(2000, random_state=42)

fig, axes = plt.subplots(1, 3, figsize=(14, 5))
# Price by cut
sns.violinplot(data=sample, x='cut', y='price', palette='Set2',
               inner='quartile', ax=axes[0])
axes[0].set_title('Price by Cut')

# Carat vs price scatter
sns.scatterplot(data=sample, x='carat', y='price', hue='color',
                palette='RdYlGn', s=15, alpha=0.5, ax=axes[1])
axes[1].set_title('Carat vs Price by Color')

# Correlation heatmap of numeric cols
corr = sample[['carat','depth','table','price','x','y','z']].corr()
sns.heatmap(corr, annot=True, fmt='.2f', cmap='coolwarm',
            center=0, linewidths=0.5, ax=axes[2])
axes[2].set_title('Feature Correlation')

fig.suptitle('Diamonds EDA', fontweight='bold', fontsize=13)
fig.tight_layout()
fig.savefig('diamonds_eda.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved diamonds_eda.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
penguins = sns.load_dataset('penguins').dropna()

fig, axes = plt.subplots(2, 2, figsize=(12, 9))
sns.scatterplot(data=penguins, x='bill_length_mm', y='bill_depth_mm',
                hue='species', style='island', s=60, alpha=0.7, ax=axes[0,0])
axes[0,0].set_title('Bill Length vs Depth by Species & Island')

sns.violinplot(data=penguins, x='species', y='body_mass_g',
               hue='sex', split=True, palette='Set2', inner='box',
               ax=axes[0,1])
axes[0,1].set_title('Body Mass by Species & Sex')

sns.ecdfplot(data=penguins, x='flipper_length_mm', hue='species',
             palette='Set1', linewidth=2, ax=axes[1,0])
axes[1,0].set_title('Flipper Length ECDF by Species')

counts = penguins.groupby(['species','island']).size().reset_index(name='count')
sns.barplot(data=counts, x='species', y='count', hue='island',
            palette='pastel', ax=axes[1,1])
axes[1,1].set_title('Population by Species & Island')

fig.suptitle('Penguins Complete EDA', fontweight='bold', fontsize=13)
fig.tight_layout()
fig.savefig('penguins_eda.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved penguins_eda.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
mpg = sns.load_dataset('mpg').dropna()

fig, axes = plt.subplots(1, 3, figsize=(14, 5))
# MPG trend by model year and origin
sns.lineplot(data=mpg, x='model_year', y='mpg', hue='origin',
             palette='Set1', linewidth=2, markers=True, ax=axes[0])
axes[0].set_title('MPG by Year & Origin')

# Weight vs MPG regression by cylinders
cyl_order = sorted(mpg['cylinders'].unique())
pal = sns.color_palette('coolwarm', len(cyl_order))
for i, cyl in enumerate(cyl_order):
    sub = mpg[mpg.cylinders == cyl]
    axes[1].scatter(sub.weight, sub.mpg, s=20, alpha=0.5, color=pal[i], label=f'{cyl}cyl')
axes[1].set_xlabel('Weight (lbs)'); axes[1].set_ylabel('MPG')
axes[1].legend(title='Cylinders', fontsize=8); axes[1].set_title('Weight vs MPG by Cylinders')

# Horsepower distribution by origin
sns.boxplot(data=mpg, x='origin', y='horsepower', palette='Set2', ax=axes[2])
axes[2].set_title('Horsepower by Origin')

fig.suptitle('MPG Dataset — Fuel Efficiency EDA', fontweight='bold', fontsize=13)
fig.tight_layout()
fig.savefig('mpg_eda.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved mpg_eda.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

import seaborn.objects as so
sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

p = (
    so.Plot(tips, x='total_bill', y='tip', color='sex')
    .add(so.Dot(alpha=0.6, pointsize=5))
    .add(so.Line(), so.PolyFit(order=1))
    .label(x='Total Bill ($)', y='Tip ($)', color='Sex',
           title='Tips: Scatter + Regression (objects API)')
)
p.save('so_scatter.png', dpi=120, bbox_inches='tight')
print('Saved so_scatter.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

import seaborn.objects as so
sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

p = (
    so.Plot(tips, x='day', y='total_bill', color='sex')
    .add(so.Bar(), so.Agg('mean'), so.Dodge())
    .label(x='Day', y='Mean Total Bill ($)', color='Sex',
           title='Mean Bill by Day & Sex (objects API)')
)
p.save('so_bar.png', dpi=120, bbox_inches='tight')
print('Saved so_bar.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

import seaborn.objects as so
sns.set_theme(style='whitegrid')
iris = sns.load_dataset('iris')

p = (
    so.Plot(iris, x='petal_length', color='species')
    .add(so.Bars(), so.Hist(binwidth=0.3), so.Norm('percent'))
    .add(so.Line(), so.KDE())
    .label(x='Petal Length (cm)', y='Percent',
           title='Petal Length Distribution (objects API)')
)
p.save('so_hist.png', dpi=120, bbox_inches='tight')
print('Saved so_hist.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

import seaborn.objects as so
sns.set_theme(style='whitegrid')
tips = sns.load_dataset('tips')

p = (
    so.Plot(tips, x='total_bill', y='tip', color='sex')
    .facet(col='time', row='smoker')
    .add(so.Dot(alpha=0.5, pointsize=4))
    .label(x='Total Bill ($)', y='Tip ($)',
           title='Tips: Faceted Grid (objects API)')
    .limit(x=(0, 55), y=(0, 11))
)
p.save('so_faceted.png', dpi=120, bbox_inches='tight')
print('Saved so_faceted.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(0)
df = pd.DataFrame({
    'value': np.random.randn(200),
    'group': np.random.choice(list('ABCDE'), 200)
})

qual_palettes = ['Set1','Set2','Set3','tab10','colorblind','deep']
fig, axes = plt.subplots(2, 3, figsize=(14, 8))
for ax, pal in zip(axes.flat, qual_palettes):
    sns.boxplot(data=df, x='group', y='value', palette=pal, ax=ax)
    ax.set_title(f'palette="{pal}"')
fig.suptitle('Qualitative Palettes Comparison', fontweight='bold', fontsize=13)
fig.tight_layout()
fig.savefig('qual_palettes.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved qual_palettes.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(1)
x = y = np.linspace(-3, 3, 60)
X, Y = np.meshgrid(x, y)
Z = np.sin(X) * np.cos(Y)

seq_palettes = ['Blues','YlOrRd','viridis','magma']
fig, axes = plt.subplots(1, 4, figsize=(16, 4))
for ax, pal in zip(axes, seq_palettes):
    im = ax.pcolormesh(X, Y, Z, cmap=pal)
    fig.colorbar(im, ax=ax, shrink=0.8)
    ax.set_title(f'cmap="{pal}"')
    ax.set_aspect('equal')
fig.suptitle('Sequential Palettes for Heatmap Data', fontweight='bold')
fig.tight_layout()
fig.savefig('seq_palettes.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved seq_palettes.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='white')
np.random.seed(2)
data = np.random.randn(10, 8) * 2
df = pd.DataFrame(data,
    index=[f'Product {i}' for i in range(1,11)],
    columns=[f'Q{q}' for q in range(1,9)])

div_palettes = ['RdBu_r','RdYlGn','coolwarm','PiYG']
fig, axes = plt.subplots(1, 4, figsize=(16, 4))
for ax, pal in zip(axes, div_palettes):
    sns.heatmap(df, cmap=pal, center=0, ax=ax,
                cbar_kws={'shrink':0.8}, xticklabels=True, yticklabels=False)
    ax.set_title(f'cmap="{pal}"')
fig.suptitle('Diverging Palettes for Change Data', fontweight='bold')
fig.tight_layout()
fig.savefig('div_palettes.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved div_palettes.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

sns.set_theme(style='whitegrid')
np.random.seed(3)
df = pd.DataFrame({
    'x': np.linspace(0, 10, 80),
    'A': np.sin(np.linspace(0, 10, 80)) + np.random.randn(80)*0.2,
    'B': np.cos(np.linspace(0, 10, 80)) + np.random.randn(80)*0.2,
    'C': np.sin(np.linspace(0, 10, 80)*1.5) + np.random.randn(80)*0.2,
})

fig, axes = plt.subplots(1, 2, figsize=(13, 5))
# Non-safe: red-green confusion
colors_bad = ['#ff0000','#00aa00','#0000ff']
for col, c in zip(['A','B','C'], colors_bad):
    axes[0].plot(df.x, df[col], color=c, linewidth=2, label=col,
                 linestyle='-')
axes[0].set_title('Non-colorblind-safe (avoid red/green)')
axes[0].legend()

# Colorblind-safe (Wong palette) with markers
cb_colors = ['#0072B2','#D55E00','#009E73']
markers = ['o','s','^']
for col, c, m in zip(['A','B','C'], cb_colors, markers):
    axes[1].plot(df.x, df[col], color=c, linewidth=2,
                 label=col, marker=m, markevery=8, markersize=7)
axes[1].set_title('Colorblind-safe (Wong palette + markers)')
axes[1].legend()

for ax in axes:
    ax.set_xlabel('x'); ax.set_ylabel('y')
    ax.grid(True, alpha=0.3)
fig.suptitle('Accessibility: Color Choice Matters', fontweight='bold')
fig.tight_layout()
fig.savefig('colorblind_safe.png', dpi=120, bbox_inches='tight')
plt.close()
print('Saved colorblind_safe.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

from matplotlib.gridspec import GridSpec

sns.set_theme(style='whitegrid', palette='Set2', font_scale=0.95)
np.random.seed(42)
df = pd.DataFrame({
    'revenue': np.random.lognormal(5, 0.5, 300),
    'spend':   np.random.uniform(1000, 10000, 300),
    'channel': np.random.choice(['Search','Social','Email','Direct'], 300),
    'month':   np.random.choice(range(1,13), 300)
})
df['revenue'] = df['spend'] * np.random.uniform(0.1, 0.5, 300) + np.random.randn(300)*50

fig = plt.figure(figsize=(14, 9))
gs = GridSpec(2, 3, figure=fig, hspace=0.4, wspace=0.35)

ax1 = fig.add_subplot(gs[0, :2])
ax2 = fig.add_subplot(gs[0, 2])
ax3 = fig.add_subplot(gs[1, :])

sns.scatterplot(data=df, x='spend', y='revenue', hue='channel',
                style='channel', s=50, alpha=0.6, ax=ax1)
ax1.set_title('Revenue vs Spend by Channel', fontweight='bold')

sns.violinplot(data=df, x='channel', y='revenue', palette='Set2',
               inner='quartile', ax=ax2)
ax2.set_title('Revenue Distribution', fontweight='bold')
ax2.set_xticklabels(ax2.get_xticklabels(), rotation=20, ha='right', fontsize=8)

month_agg = df.groupby(['month','channel'])['revenue'].mean().reset_index()
sns.lineplot(data=month_agg, x='month', y='revenue', hue='channel',
             palette='Set2', linewidth=2, markers=True, ax=ax3)
ax3.set_title('Monthly Average Revenue by Channel', fontweight='bold')
ax3.set_xlabel('Month')

fig.suptitle('Marketing Revenue Dashboard', fontsize=14, fontweight='bold')
fig.savefig('sns_dashboard.png', dpi=150, bbox_inches='tight')
plt.close()
print('Saved sns_dashboard.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

from scipy import stats

sns.set_theme(style='whitegrid')
np.random.seed(1)
control    = np.random.normal(70, 12, 100)
treatment  = np.random.normal(75, 11, 100)
t_stat, p_val = stats.ttest_ind(control, treatment)
effect_size = (treatment.mean() - control.mean()) / np.sqrt(
    ((len(control)-1)*control.std()**2 + (len(treatment)-1)*treatment.std()**2) /
    (len(control)+len(treatment)-2))

df = pd.DataFrame({
    'score': np.concatenate([control, treatment]),
    'group': np.repeat(['Control','Treatment'], 100)
})

fig, axes = plt.subplots(1, 3, figsize=(14, 5))
sns.histplot(data=df, x='score', hue='group', kde=True, alpha=0.4,
             palette='Set1', stat='density', common_norm=False, ax=axes[0])
axes[0].set_title('Score Distributions')

sns.violinplot(data=df, x='group', y='score', palette='Set1',
               inner='box', ax=axes[1])
axes[1].set_title('Violin Comparison')

sns.ecdfplot(data=df, x='score', hue='group', palette='Set1',
             linewidth=2, ax=axes[2])
axes[2].set_title('ECDF Comparison')

for ax in axes:
    ax.set_xlabel('Score')

fig.suptitle(f'A/B Test: t={t_stat:.2f}, p={p_val:.4f}, d={effect_size:.2f}',
             fontsize=13, fontweight='bold')
fig.tight_layout()
fig.savefig('hypothesis_dashboard.png', dpi=150, bbox_inches='tight')
plt.close()
print('Saved hypothesis_dashboard.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

from matplotlib.gridspec import GridSpec

sns.set_theme(style='whitegrid', palette='colorblind')
penguins = sns.load_dataset('penguins').dropna()

fig = plt.figure(figsize=(15, 10), facecolor='white')
gs = GridSpec(3, 3, figure=fig, hspace=0.5, wspace=0.4)

# Banner
ax0 = fig.add_subplot(gs[0, :])
ax0.text(0.5, 0.6, 'Palmer Penguins — Deep Dive EDA',
         ha='center', va='center', fontsize=16, fontweight='bold',
         transform=ax0.transAxes)
ax0.text(0.5, 0.1, f'n={len(penguins)} penguins | 3 species | 3 islands',
         ha='center', va='center', fontsize=11, color='gray',
         transform=ax0.transAxes)
ax0.axis('off')

# Scatter bill dimensions
ax1 = fig.add_subplot(gs[1, :2])
sns.scatterplot(data=penguins, x='bill_length_mm', y='bill_depth_mm',
                hue='species', style='island', s=60, alpha=0.7, ax=ax1)
ax1.set_title('Bill Dimensions by Species & Island', fontweight='bold')

# Count by species
ax2 = fig.add_subplot(gs[1, 2])
sns.countplot(data=penguins, x='species', hue='sex',
              palette='Set2', ax=ax2)
ax2.set_title('Counts by Species & Sex', fontweight='bold')
ax2.set_xlabel('Species')

# Body mass violin
ax3 = fig.add_subplot(gs[2, :2])
sns.violinplot(data=penguins, x='species', y='body_mass_g',
               hue='sex', split=True, palette='pastel', inner='box', ax=ax3)
ax3.set_title('Body Mass Distribution', fontweight='bold')

# Flipper ECDF
ax4 = fig.add_subplot(gs[2, 2])
sns.ecdfplot(data=penguins, x='flipper_length_mm', hue='species',
             palette='colorblind', linewidth=2, ax=ax4)
ax4.set_title('Flipper Length ECDF', fontweight='bold')
ax4.set_xlabel('Flipper Length (mm)')

fig.savefig('penguins_deep_dive.png', dpi=150, bbox_inches='tight',
            facecolor=fig.get_facecolor())
plt.close()
print('Saved penguins_deep_dive.png')

import matplotlib
matplotlib.use('Agg')
import seaborn as sns
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

from matplotlib.gridspec import GridSpec

sns.set_theme(style='darkgrid', palette='Set2')
np.random.seed(77)
n_days = 90
dates = pd.date_range('2024-01-01', periods=n_days)
products = ['Widget','Gadget','Doohickey']
rows = []
for prod in products:
    base = np.random.uniform(80, 150)
    trend = np.cumsum(np.random.randn(n_days)*2) + base
    for d, v in zip(dates, trend):
        rows.append({'date':d,'product':prod,'sales':max(0,v),
                     'returns':max(0,v*0.05+np.random.randn())})
df = pd.DataFrame(rows)
df['month'] = df['date'].dt.month

fig = plt.figure(figsize=(14, 9))
gs = GridSpec(2, 2, figure=fig, hspace=0.4, wspace=0.35)

ax1 = fig.add_subplot(gs[0, :])
sns.lineplot(data=df, x='date', y='sales', hue='product',
             palette='Set2', linewidth=2, ax=ax1)
ax1.set_title('Daily Sales Trend by Product', fontweight='bold')
ax1.set_xlabel('Date'); ax1.set_ylabel('Sales ($)')

ax2 = fig.add_subplot(gs[1, 0])
month_df = df.groupby(['month','product'])['sales'].mean().reset_index()
sns.barplot(data=month_df, x='month', y='sales', hue='product',
            palette='Set2', ax=ax2)
ax2.set_title('Mean Monthly Sales', fontweight='bold')
ax2.set_xlabel('Month')

ax3 = fig.add_subplot(gs[1, 1])
sns.scatterplot(data=df, x='sales', y='returns', hue='product',
                palette='Set2', s=20, alpha=0.4, ax=ax3)
# Add regression per product
for prod, grp in df.groupby('product'):
    m,b = np.polyfit(grp.sales, grp.returns, 1)
    x_l = np.linspace(grp.sales.min(), grp.sales.max(), 50)
    ax3.plot(x_l, m*x_l+b, linewidth=1.5, alpha=0.8)
ax3.set_title('Sales vs Returns', fontweight='bold')

fig.suptitle('Product Analytics Dashboard', fontsize=14, fontweight='bold')
fig.savefig('time_series_dashboard.png', dpi=150, bbox_inches='tight')
plt.close()
print('Saved time_series_dashboard.png')