Skip to main content

A package for visualizing state transition graphs from hidden Markov models or other models.

Project description


A package for visualizing state transition graphs from hidden Markov models or other models.

This package is meant to serve as an out-of-the-box means of plotting small graphs (less than 10 nodes) in a customizable way.


Developed with Python 3.6

pip install hmmviz


Each graph object takes a pandas DataFrame with states indices and columns and transition probabilities (from index to column). However, these can be instantiated using different data structures such as numpy arrays, networkx graphs, etc. (help needed) via the appropriate class methods. as values.

Plotting a Transition Matrix from a Markov Process

from hmmviz import TransGraph
import pandas as pd
import matplotlib.pyplot as plt

sequence = ['sunny', 'sunny', 'rainy', 'rainy', 'sunny', 'sunny', 'sunny', 'rainy']

T = pd.crosstab(
    pd.Series(sequence[:-1], name='Today'),
    pd.Series(sequence[1:], name='Tomorrow'),
# Tomorrow  rainy  sunny
# Today                 
# rainy       0.5    0.5
# sunny       0.4    0.6

graph = TransGraph(T)

# looks best on square figures/axes
fig = plt.figure(figsize=(6, 6))


This will make an all black graph with labels on the nodes but none on the edges.

If we want to make the graph more colorful and informative we can pass some parameters into the graph's draw method.

# same T as before
graph = TransGraph(T)

fig = plt.figure(figsize=(6, 6))

nodelabels = {'sunny':  '☁☀', 'rainy': '😊☂'}
colors = {'sunny': 'orange', 'rainy': 'blue'}

    nodelabels=nodelabels, nodecolors=colors, edgecolors=colors, edgelabels=True,

hmmviz really shines for graphs with 4 or 5 nodes. For larger graphs consider using a package like nxviz.

import numpy as np

arr = np.ones((4, 4)) * 0.25

labels = ['s0', 's1', 's2', 's3']

graph = TransGraph.from_array(arr, labels)

colors = {f's{i}': f'C{i}' for i in range(4)}

plt.figure(figsize=(6, 6))

graph.draw(edgecolors=colors, nodecolors=colors, nodelabels=False, edgewidths=2)

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for hmmviz, version 0.0.7
Filename, size File type Python version Upload date Hashes
Filename, size hmmviz-0.0.7.tar.gz (5.4 kB) File type Source Python version None Upload date Hashes View
Filename, size hmmviz-0.0.7-py3-none-any.whl (6.1 kB) File type Wheel Python version py3 Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page