og

Frequent itemsets -- fp-tree naeseth

To install: pip install og

Overview

The og package provides a Python implementation of the FP-growth algorithm for finding frequent itemsets in transactional data sets. This method is efficient and scalable, making it suitable for large data sets where traditional apriori-based methods may be too slow. It constructs a compressed representation of the dataset, the FP-tree, which is then used to extract frequent itemsets directly.

Features

• Efficiently find frequent itemsets without candidate generation.
• Return itemsets along with their support counts if desired.
• Handle any iterable of iterables as input for transactions.

Usage

Basic Usage

To use the find_frequent_itemsets function, you need to provide a list of transactions and a minimum support threshold. Here is a simple example:

from og import find_frequent_itemsets

# Sample transactions
transactions = [
    ['milk', 'bread', 'butter'],
    ['beer', 'bread'],
    ['milk', 'bread'],
    ['butter', 'beer'],
    ['bread', 'butter'],
    ['milk', 'butter'],
    ['milk', 'bread', 'butter', 'beer']
]

# Finding itemsets with a minimum support of 3
result = list(find_frequent_itemsets(transactions, minimum_support=3))
print(result)

Including Support Counts

If you want to include the support counts of the itemsets in the results, set include_support=True:

# Finding itemsets with support counts
result_with_support = list(find_frequent_itemsets(transactions, minimum_support=3, include_support=True))
print(result_with_support)

Classes and Functions

FPTree

This class represents an FP-tree structure that can store transaction items. It supports adding transactions, and provides methods to access the nodes and items.

FPNode

Represents a node in an FP-tree. Each node contains a count of occurrences, links to child nodes, and a reference to its parent node.

find_frequent_itemsets

A function to find frequent itemsets in the given transactions using the FP-growth algorithm. It can optionally return the support count for each itemset.

Parameters:

• transactions: An iterable of iterable items. Each inner iterable represents a transaction.
• minimum_support: An integer specifying the minimum number of occurrences for an itemset to be considered frequent.
• include_support: A boolean flag that, when set to True, includes the support count of each itemset in the results.

Development and Contributions

Contributions to the og package are welcome. You can contribute in several ways including providing feedback, reporting bugs, and submitting feature requests or pull requests.

For more detailed contribution guidelines, please refer to the official repository documentation.

License

This project is licensed under the MIT License - see the LICENSE file for details.