Histogram Plot Gallery¶

The histogram plot is used for visualizing the distribution of numerical data by grouping values into bins and showing frequencies. It's ideal for understanding data patterns, distributions, and identifying outliers or trends in your retail analytics.

Histogram plots excel at:

• Distribution Analysis: Visualize the spread and shape of key metrics like sales amounts or customer spending
• Group Comparisons: Compare distributions across categories like product types or customer segments
• Range Analysis: Focus on specific value ranges using clipping or filtering options
• Outlier Detection: Identify unusual patterns or extreme values in your data

In [ ]:

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd

from pyretailscience.plots import histogram

import matplotlib.pyplot as plt import numpy as np import pandas as pd from pyretailscience.plots import histogram

Basic Histogram¶

Create a simple histogram from a single value column to visualize the distribution of your data.

---

In [ ]:

# Create sample transaction amounts data
rng = np.random.default_rng(42)
transaction_data = pd.DataFrame(
    {
        "transaction_amount": rng.gamma(2, 50, 500),
    },
)

histogram.plot(
    transaction_data,
    value_col="transaction_amount",
    title="Distribution of Transaction Amounts",
    x_label="Transaction Amount ($)",
    y_label="Frequency",
    bins=30,
)
plt.show()

# Create sample transaction amounts data rng = np.random.default_rng(42) transaction_data = pd.DataFrame( { "transaction_amount": rng.gamma(2, 50, 500), }, ) histogram.plot( transaction_data, value_col="transaction_amount", title="Distribution of Transaction Amounts", x_label="Transaction Amount ($)", y_label="Frequency", bins=30, ) plt.show()

Multiple Histograms with group_col¶

Compare distributions across different categories using the group_col parameter. This creates separate overlaid histograms for each group.

---

In [ ]:

# Create sample data with customer segments
rng = np.random.default_rng(42)
segment_data = pd.DataFrame(
    {
        "customer_id": range(1, 401),
        "order_value": np.concatenate(
            [
                rng.gamma(1.5, 30, 150),  # Bronze customers
                rng.gamma(2.5, 45, 150),  # Silver customers
                rng.gamma(3.5, 65, 100),  # Gold customers
            ],
        ),
        "segment": ["Bronze"] * 150 + ["Silver"] * 150 + ["Gold"] * 100,
    },
)

histogram.plot(
    segment_data,
    value_col="order_value",
    group_col="segment",
    title="Order Value Distribution by Customer Segment",
    x_label="Order Value ($)",
    y_label="Frequency",
    legend_title="Customer Segment",
    move_legend_outside=True,
    bins=25,
)
plt.show()

# Create sample data with customer segments rng = np.random.default_rng(42) segment_data = pd.DataFrame( { "customer_id": range(1, 401), "order_value": np.concatenate( [ rng.gamma(1.5, 30, 150), # Bronze customers rng.gamma(2.5, 45, 150), # Silver customers rng.gamma(3.5, 65, 100), # Gold customers ], ), "segment": ["Bronze"] * 150 + ["Silver"] * 150 + ["Gold"] * 100, }, ) histogram.plot( segment_data, value_col="order_value", group_col="segment", title="Order Value Distribution by Customer Segment", x_label="Order Value ($)", y_label="Frequency", legend_title="Customer Segment", move_legend_outside=True, bins=25, ) plt.show()

Multiple Value Columns¶

Plot multiple metrics simultaneously by passing a list to value_col. Note: This cannot be combined with group_col.

---

In [ ]:

# Create sample data with multiple metrics
rng = np.random.default_rng(42)
metrics_data = pd.DataFrame(
    {
        "Online Revenue": rng.gamma(2, 100, 300),
        "Store Revenue": rng.gamma(2.5, 120, 300),
        "Mobile Revenue": rng.gamma(1.8, 80, 300),
    },
)

histogram.plot(
    metrics_data,
    value_col=["Online Revenue", "Store Revenue", "Mobile Revenue"],
    title="Revenue Distribution by Sales Channel",
    x_label="Daily Revenue ($)",
    y_label="Frequency",
    legend_title="Sales Channel",
    move_legend_outside=True,
    bins=20,
)
plt.show()

# Create sample data with multiple metrics rng = np.random.default_rng(42) metrics_data = pd.DataFrame( { "Online Revenue": rng.gamma(2, 100, 300), "Store Revenue": rng.gamma(2.5, 120, 300), "Mobile Revenue": rng.gamma(1.8, 80, 300), }, ) histogram.plot( metrics_data, value_col=["Online Revenue", "Store Revenue", "Mobile Revenue"], title="Revenue Distribution by Sales Channel", x_label="Daily Revenue ($)", y_label="Frequency", legend_title="Sales Channel", move_legend_outside=True, bins=20, ) plt.show()