Missingno

Common errors

• AttributeError: module 'missingno' has no attribute 'geoplot'

  cause The `geoplot` method was removed in `missingno` version 0.5.0.
  fix Remove the call to `msno.geoplot()`. For geospatial data, use a dedicated library like `geopandas`.

• TypeError: matrix() got an unexpected keyword argument 'inline'

  cause The `inline` parameter was removed from `missingno` visualization functions in version 0.5.0.
  fix Remove the `inline=True` or `inline=False` argument from your `missingno` function calls.

• TypeError: bar() got an unexpected keyword argument 'sort'

  cause The `sort` parameter's functionality and acceptance for certain plot types like `dendrogram` (and implicitly `bar` in older contexts) changed or was removed in `missingno` version 0.4.2.
  fix Check the documentation for the specific `missingno` plot function. For `dendrogram`, remove the `sort` parameter entirely. For `bar`, ensure your `sort` parameter aligns with current valid options (e.g., 'ascending', 'descending').

• ValueError: Invalid 'kind' argument for plot_nullity. Must be one of ['matrix', 'bar', 'heatmap', 'dendrogram'].

  cause Attempting to call a non-existent or misspelled plot kind, possibly from older documentation or misremembered functionality.
  fix Ensure the plot function name is correct (e.g., `msno.matrix()`, `msno.bar()`, `msno.heatmap()`, `msno.dendrogram()`). The library does not have a generic `plot_nullity` function with a `kind` argument.

Warnings

• breaking The `geoplot` method and the `inline` parameter for visualizations were removed in `missingno` version 0.5.0. Code relying on these will break.
  Fix: Remove calls to `msno.geoplot()` and the `inline` parameter. For geospatial analysis, consider using `geopandas` directly or an alternative visualization.
• deprecated The `sort` parameter's behavior changed significantly and was removed from `dendrogram` and `geoplot` in version 0.4.2. Using it in older ways for these plots will not work as expected or raise errors in newer versions.
  Fix: Review `missingno` documentation for current `sort` parameter usage. If specifically using `dendrogram`, remove the `sort` parameter as it's no longer supported.
• gotcha Older versions of `missingno` (prior to 0.5.2) may experience compatibility issues with newer versions of `matplotlib`, leading to visual glitches or errors.
  Fix: Upgrade `missingno` to version 0.5.2 or later: `pip install --upgrade missingno`.
• gotcha When using the `ax` parameter to plot `msno.matrix` onto a `matplotlib.axes.Axes` object, the `sparkline` parameter is not supported and will be ignored or cause issues.
  Fix: Avoid using the `sparkline` parameter when passing an `ax` object to `msno.matrix`.
• gotcha Missingno is primarily designed to work with Pandas DataFrames. Attempting to use it directly with other data structures (e.g., raw NumPy arrays or lists) will require conversion to a DataFrame first.
  Fix: Ensure your data is a Pandas DataFrame before passing it to `missingno` functions.

Install

• pip install missingno Install stable version

Imports

• missingno

  import missingno as msno

• pandas

  import pandas as pd

• numpy

  import numpy as np

Quickstart

This quickstart demonstrates how to create a Pandas DataFrame with simulated missing values and then visualize them using `missingno.matrix` and `missingno.bar`. The matrix plot provides a visual summary of missing data patterns, while the bar chart shows the count of non-null values per column.

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

# Create a sample DataFrame with missing values
data = {
    'A': [1, 2, np.nan, 4, 5],
    'B': [np.nan, 2, 3, 4, np.nan],
    'C': [1, 2, 3, np.nan, 5],
    'D': [1, 2, 3, 4, 5]
}
df = pd.DataFrame(data)

print("DataFrame with missing values:")
print(df)
print("\nMissingno Matrix Visualization:")

# Generate a missingness matrix plot
msno.matrix(df, figsize=(8, 4))
plt.title('Missing Data Matrix')
plt.show()

print("\nMissingno Bar Chart Visualization:")
# Generate a bar chart of missingness
msno.bar(df, figsize=(8, 4))
plt.title('Missing Data Bar Chart')
plt.show()