EDA & Visualization

# Walmart Logistics: Outlier Detection (Cleaning)
class LogisticsCleaner:
    def __init__(self, transit_times):
        self.times = transit_times # List of hours in transit

    def remove_anomalies(self, threshold=30):
        print("CLEANING: Identifying transit outliers > {0} hours...".format(threshold))
        clean = [t for t in self.times if t <= threshold]
        anomalies = [t for t in self.times if t > threshold]
        
        print(" - Found {0} anomalies: {1}".format(len(anomalies), anomalies))
        print(" - Clean Dataset Size: {0} routes".format(len(clean)))
        return clean

# Scenario: Auditing the Mumbai-Jaipur route
raw_times = [18, 19, 21, 150, 20, 185, 22] # 150/185 are clear errors/crashes
cleaner = LogisticsCleaner(raw_times)
clean_data = cleaner.remove_anomalies()

# Walmart Supply Chain: Correlation EDA
class RouteExplorer:
    def __init__(self, data_points):
        self.data = data_points # List of (Temp, SpoilageRate)

    def analyze_correlation(self):
        print("EDA: Analyzing Temp vs. Spoilage correlation...")
        # Simulating a basic trend analysis
        increasing = all(self.data[i][1] <= self.data[i+1][1] for i in range(len(self.data)-1))
        
        for temp, rate in self.data:
            print(" - Temp: {0}°C -> Spoilage: {1:.1f}%".format(temp, rate))
        
        if increasing:
            print("\nINSIGHT: Strong Positive Correlation detected. Increasing temp correlates with higher spoilage.")

# Scenario: Monitoring refrigerated containers
sensor_data = [(4, 0.1), (6, 0.5), (10, 2.5), (15, 8.0)]
explorer = RouteExplorer(sensor_data)
explorer.analyze_correlation()

# Walmart Visualizer: Heatmap Simulation
class BottleneckVisualizer:
    def __init__(self, grid_data):
        self.grid = grid_data # 2D List of capacity % [Region A, Region B]

    def render_heatmap(self):
        print("VISUALIZATION: Generating Capacity Heatmap (Seaborn-style)...")
        symbols = {0: "🟢", 50: "🟡", 80: "🟠", 95: "🔴"}
        
        for row_idx, row in enumerate(self.grid):
            line = "Region {0}: ".format(chr(65 + row_idx))
            for val in row:
                # Find the closest status symbol
                status = symbols[max(k for k in symbols if k <= val)]
                line += "{0} {1}%  ".format(status, val)
            print(line)
        
        print("\nGUIDE: 🔴 > 95% (CRITICAL) | 🟢 < 50% (AVAILABLE)")

# Scenario: Global Warehouse Monitoring
global_load = [
    [40, 42, 98], # Region A (Warehouse 3 is Critical)
    [55, 60, 52]  # Region B
]
viz = BottleneckVisualizer(global_load)
viz.render_heatmap()

# Matplotlib: 20 Essential Methods (Part 1)
class MatplotlibReference:
    def demonstrate(self):
        methods = [
            ("1. plt.figure()", "Create a new figure canvas", "Walmart: Initialize a new dashboard panel"),
            ("2. plt.subplot()", "Add multiple charts in a grid", "Walmart: Side-by-side regional comparison"),
            ("3. plt.plot()", "Draw a line chart", "Netflix: Subscriber growth over time"),
            ("4. plt.bar()", "Draw a bar chart", "Walmart: Revenue by warehouse region"),
            ("5. plt.barh()", "Horizontal bar chart", "Visa: Top 10 merchants by volume"),
            ("6. plt.scatter()", "Draw a scatter plot", "Netflix: Watch time vs. user rating"),
            ("7. plt.hist()", "Draw a histogram", "Visa: Distribution of transaction amounts"),
            ("8. plt.pie()", "Draw a pie chart", "Walmart: Market share by product category"),
            ("9. plt.xlabel() / ylabel()", "Label the axes", "All: Context for every chart"),
            ("10. plt.title()", "Set the chart title", "All: Clear headline for stakeholders")
        ]
        
        print("MATPLOTLIB REFERENCE (Part 1: Core Drawing):")
        print("=" * 55)
        for name, desc, scenario in methods:
            print("{0}".format(name))
            print("  What: {0}".format(desc))
            print("  Use : {0}".format(scenario))
            print("")

ref = MatplotlibReference()
ref.demonstrate()

# Seaborn & Matplotlib: 20 Essential Methods (Part 2)
class SeabornReference:
    def demonstrate(self):
        methods = [
            ("11. plt.legend()", "Add a legend to the chart", "Netflix: Differentiate genre lines"),
            ("12. plt.xlim() / ylim()", "Set axis range limits", "Visa: Zoom into $0-$500 range"),
            ("13. plt.grid()", "Add gridlines for readability", "Walmart: Align visual to values"),
            ("14. plt.savefig()", "Export chart as PNG/PDF", "All: Embed in executive reports"),
            ("15. plt.tight_layout()", "Auto-fix overlapping labels", "All: Clean multi-panel dashboards"),
            ("16. sns.heatmap()", "Correlation matrix heatmap", "Visa: Feature correlation in fraud model"),
            ("17. sns.boxplot()", "Show distribution + outliers", "Walmart: Transit time spread per route"),
            ("18. sns.violinplot()", "Distribution shape + density", "Netflix: Watch time distribution by genre"),
            ("19. sns.pairplot()", "All-vs-all scatter matrix", "Visa: Explore relationships in 200 features"),
            ("20. sns.countplot()", "Count frequency of categories", "Netflix: Most popular genres bar chart")
        ]
        
        print("SEABORN & MATPLOTLIB REFERENCE (Part 2: Advanced):")
        print("=" * 55)
        for name, desc, scenario in methods:
            print("{0}".format(name))
            print("  What: {0}".format(desc))
            print("  Use : {0}".format(scenario))
            print("")

ref = SeabornReference()
ref.demonstrate()