chonyy/fpgrowth_py

🔨 Python implementation of FP Growth algorithm, new and simple!

✏️ Latest Articles

• 📕 SimRank: Similarity Analysis Explanation and Python Implementation from Scratch
• 📗 PageRank: Link Analysis Explanation and Python Implementation from Scratch
• 📘 HITS Algorithm: Link Analysis Explanation and Python Implementation from Scratch
• 📙 FP Growth: Frequent Pattern Generation in Data Mining with Python Implementation

👨‍💻 Working on

• Neo4jCsvPublisher Speed Optimization (Parallelism)
• Add postgres compatibility in HiveTableLastUpdatedExtractor
• Neo4j csv publisher using apoc library for performance improvements

Note: This readme file is generated by CI/CD using GitHub Actions. Automatically updated every hour!

fpgrowth_py's Issues

Add support ratio in the fpgrowth's return rules

Your project really help me. But could you please add support ratio in the fpgrowth function's return rules?

TypeError: cannot unpack non-iterable NoneType object

from fpgrowth_py import fpgrowth
itemSetList = [['276', '303'],
 ['0', '109', '293', '160', '171'],
 ['204', '49'],
 ['204', '105'],
 ['49', '303', '105', '204', '276'],
 ['261', '255', '154', '74'],
 ['145', '261', '154']]
freqItemSet, rules = fpgrowth(itemSetList, minSupRatio=0.5, minConf=0.5)

TypeError: cannot unpack non-iterable NoneType object

ModuleNotFoundError

hi, run the pip installation, but still gives me "ModuleNotFoundError: No module named 'fpgrowth_py'".
Any help? Thanks!

The output is wrong when the minConf=1

from fpgrowth_py import fpgrowth
itemSetList = [['eggs', 'bacon', 'soup'],
               ['eggs', 'bacon', 'apple'],
               ['soup', 'bacon', 'banana']]
freqItemSet, rules = fpgrowth(itemSetList, minSupRatio=0.5, minConf=1)
print(rules)

THE EXPECTED OUTPUT RULES SHOULD BE:
[{eggs} -> {bacon}, {soup} -> {bacon}]

BUT THE OUTPUT OF THE ABOVE CODE IS:
[]

this error is found when comparing your package with another package efficient_apriori (see https://github.com/tommyod/Efficient-Apriori)

I need the support of each frequent pattern.

I need the support of each frequent pattern. can you just tell me how to get every pattern's support with it? sth like this:
[{'L' : 4}, {'L', 'M' : 3} , ...]