Click here for more information

---

Table of Contents

• Introduction
• Development process
• Inputs and outputs of a PAMI algorithm
• Recent updates
• Features
• Maintenance
• Try your first PAMI program
• Evaluation
• Reading Material
• License
• Documentation
• Background
• Getting Help
• Discussion and Development
• Contribution to PAMI
• Tutorials
  • Association rule mining
  • Mining transactional databases
  • Mining temporal databases
  • Mining spatiotemporal databases
  • Mining utility databases
  • Mining fuzzy databases
  • Mining uncertain databases
  • Mining sequence databases
  • Mining multiple timeseries
  • Mining streams
  • Mining character sequences
  • Mining graphs
  • Additional features
    • Synthetic data generator
    • Dataframes to databases
    • Gathering database statistics
• Real-World Case Studies

---

Introduction

PAttern MIning (PAMI) is a Python library containing several algorithms to discover user interest-based patterns in a wide-spectrum of datasets across multiple computing platforms. Useful links to utilize the services of this library were provided below:

1. Youtube tutorial https://www.youtube.com/playlist?list=PLKP768gjVJmDer6MajaLbwtfC9ULVuaCZ

2. Tutorials (Notebooks) https://github.com/UdayLab/PAMI/tree/main/notebooks

3. User manual https://udaylab.github.io/PAMI/manuals/index.html

4. Coders manual https://udaylab.github.io/PAMI/codersManual/index.html

5. Code documentation https://pami-1.readthedocs.io

6. Datasets https://u-aizu.ac.jp/~udayrage/datasets.html

7. Discussions on PAMI usage https://github.com/UdayLab/PAMI/discussions

8. Report issues https://github.com/UdayLab/PAMI/issues

---

Flow Chart of Developing Algorithms in PAMI

---

Inputs and Outputs of an Algorithm in PAMI

---

Recent Updates

• Version 2024.05.01: In this latest version, the following updates have been made:
  • Included two new algorithms, Gspan and TKG, for frequent subgraph mining.
  • Updated three Synthetic Data Generator, transactional database, temporal database, and geo-referenced transactional database.
  • Optimized the following frequent pattern mining algorithms: Apriori, Aprioribitset, ECLAT, ECLATbitset, FPGrowth, and CHARM.
  • startMine() function has been deprecated to mine() function.

Total number of algorithms: 89

---

Features

• ✅ Tested to the best of our possibility
• 🔋 Highly optimized to our best effort, light-weight, and energy-efficient
• 👀 Proper code documentation
• 🍼 Ample examples of using various algorithms at ./notebooks folder
• 🤖 Works with AI libraries such as TensorFlow, PyTorch, and sklearn.
• ⚡️ Supports Cuda and PySpark
• 🖥️ Operating System Independence
• 🔬 Knowledge discovery in static data and streams
• 🐎 Snappy
• 🐻 Ease of use

---

Maintenance

Installation

1. Installing basic pami package (recommended)

   pip install pami
   

2. Installing pami package in a GPU machine that supports CUDA

   pip install 'pami[gpu]'
   

3. Installing pami package in a distributed network environment supporting Spark

   pip install 'pami[spark]'
   

4. Installing pami package for developing purpose

   pip install 'pami[dev]'
   

5. Installing complete Library of pami

   pip install 'pami[all]'
   

Upgradation

    pip install --upgrade pami

Uninstallation

    pip uninstall pami 

Information

    pip show pami

---

Try your first PAMI program

$ python

# first import pami 
from PAMI.frequentPattern.basic import FPGrowth as alg
fileURL = "https://u-aizu.ac.jp/~udayrage/datasets/transactionalDatabases/Transactional_T10I4D100K.csv"
minSup=300
obj = alg.FPGrowth(iFile=fileURL, minSup=minSup, sep='\t')
#obj.startMine()  #deprecated
obj.mine()
obj.save('frequentPatternsAtMinSupCount300.txt')
frequentPatternsDF= obj.getPatternsAsDataFrame()
print('Total No of patterns: ' + str(len(frequentPatternsDF))) #print the total number of patterns
print('Runtime: ' + str(obj.getRuntime())) #measure the runtime
print('Memory (RSS): ' + str(obj.getMemoryRSS()))
print('Memory (USS): ' + str(obj.getMemoryUSS()))

Output:
Frequent patterns were generated successfully using frequentPatternGrowth algorithm
Total No of patterns: 4540
Runtime: 8.749667644500732
Memory (RSS): 522911744
Memory (USS): 475353088

---

Evaluation:

1. we compared three different Python libraries such as PAMI, mlxtend and efficient-apriori for Apriori.
2. (Transactional_T10I4D100K.csv)is a transactional database downloaded from PAMI and used as an input file for all libraries.
3. Minimum support values and seperator are also same.

• The performance of the Apriori algorithm is shown in the graphical results below:

1. Comparing the Patterns Generated by different Python libraries for the Apriori algorithm:

   

2. Evaluating the Runtime of the Apriori algorithm across different Python libraries:

   

3. Comparing the Memory Consumption of the Apriori algorithm across different Python libraries:

   

For more information, we have uploaded the evaluation file in two formats:

• One ipynb file format, please check it here. Evaluation File ipynb
• Two pdf file format, check here. Evaluation File Pdf

---

Reading Material

For more examples, refer this YouTube link YouTube

---

License

---

Documentation

The official documentation is hosted on PAMI.

---

Background

The idea and motivation to develop PAMI was from Kitsuregawa Lab at the University of Tokyo. Work on PAMI started at University of Aizu in 2020 and has been under active development since then.

---

Getting Help

For any queries, the best place to go to is Github Issues GithubIssues.

---

Discussion and Development

In our GitHub repository, the primary platform for discussing development-related matters is the university lab. We encourage our team members and contributors to utilize this platform for a wide range of discussions, including bug reports, feature requests, design decisions, and implementation details.

---

Contribution to PAMI

We invite and encourage all community members to contribute, report bugs, fix bugs, enhance documentation, propose improvements, and share their creative ideas.

---

Tutorials

0. Association Rule Mining

Basic
Confidence
Lift
Leverage

1. Pattern mining in binary transactional databases

1.1. Frequent pattern mining: Sample

Basic	Closed	Maximal	Top-k	CUDA	pyspark
Apriori	CHARM	maxFP-growth	FAE	cudaAprioriGCT	parallelApriori
FP-growth				cudaAprioriTID	parallelFPGrowth
ECLAT				cudaEclatGCT	parallelECLAT
ECLAT-bitSet
ECLAT-diffset

1.2. Relative frequent pattern mining: Sample

Basic
RSFP-growth

1.3. Frequent pattern with multiple minimum support: Sample

Basic
CFPGrowth
CFPGrowth++

1.4. Correlated pattern mining: Sample

Basic
CoMine
CoMine++

1.5. Fault-tolerant frequent pattern mining (under development)

Basic
FTApriori
FTFPGrowth (under development)

1.6. Coverage pattern mining (under development)

Basic
CMine
CMine++

2. Pattern mining in binary temporal databases

2.1. Periodic-frequent pattern mining: Sample

Basic	Closed	Maximal	Top-K
PFP-growth	CPFP	maxPF-growth	kPFPMiner
PFP-growth++		Topk-PFP
PS-growth
PFP-ECLAT
PFPM-Compliments

2.2. Local periodic pattern mining: Sample

Basic
LPPGrowth (under development)
LPPMBreadth (under development)
LPPMDepth (under development)

2.3. Partial periodic-frequent pattern mining: Sample

Basic
GPF-growth
PPF-DFS
GPPF-DFS

2.4. Partial periodic pattern mining: Sample

Basic	Closed	Maximal	topK	CUDA
3P-growth	3P-close	max3P-growth	topK-3P growth	cuGPPMiner (under development)
3P-ECLAT				gPPMiner (under development)
G3P-Growth

2.5. Periodic correlated pattern mining: Sample

Basic
EPCP-growth

2.6. Stable periodic pattern mining: Sample

Basic	TopK
SPP-growth	TSPIN
SPP-ECLAT

2.7. Recurring pattern mining: Sample

Basic
RPgrowth

3. Mining patterns from binary Geo-referenced (or spatiotemporal) databases

3.1. Geo-referenced frequent pattern mining: Sample

Basic
spatialECLAT
FSP-growth

3.2. Geo-referenced periodic frequent pattern mining: Sample

Basic
GPFPMiner
PFS-ECLAT
ST-ECLAT

3.3. Geo-referenced partial periodic pattern mining:Sample

Basic
STECLAT

4. Mining patterns from Utility (or non-binary) databases

4.1. High utility pattern mining: Sample

Basic
EFIM
HMiner
UPGrowth

4.2. High utility frequent pattern mining: Sample

Basic
HUFIM

4.3. High utility geo-referenced frequent pattern mining: Sample

Basic
SHUFIM

4.4. High utility spatial pattern mining: Sample

Basic	topk
HDSHIM	TKSHUIM
SHUIM

4.5. Relative High utility pattern mining: Sample

Basic
RHUIM

4.6. Weighted frequent pattern mining: Sample

Basic
WFIM

4.7. Weighted frequent regular pattern mining: Sample

Basic
WFRIMiner

4.8. Weighted frequent neighbourhood pattern mining: Sample

Basic
SSWFPGrowth

5. Mining patterns from fuzzy transactional/temporal/geo-referenced databases

5.1. Fuzzy Frequent pattern mining: Sample

Basic
FFI-Miner

5.2. Fuzzy correlated pattern mining: Sample

Basic
FCP-growth

5.3. Fuzzy geo-referenced frequent pattern mining: Sample

Basic
FFSP-Miner

5.4. Fuzzy periodic frequent pattern mining: Sample

Basic
FPFP-Miner

5.5. Fuzzy geo-referenced periodic frequent pattern mining: Sample

Basic
FGPFP-Miner (under development)

6. Mining patterns from uncertain transactional/temporal/geo-referenced databases

6.1. Uncertain frequent pattern mining: Sample

Basic	top-k
PUF	TUFP
TubeP
TubeS
UVEclat

6.2. Uncertain periodic frequent pattern mining: Sample

Basic
UPFP-growth
UPFP-growth++

6.3. Uncertain Weighted frequent pattern mining: Sample

Basic
WUFIM

7. Mining patterns from sequence databases

7.1. Sequence frequent pattern mining: Sample

Basic
SPADE
PrefixSpan

7.2. Geo-referenced Frequent Sequence Pattern mining

Basic
GFSP-Miner (under development)

8. Mining patterns from multiple timeseries databases

8.1. Partial periodic pattern mining (under development)

Basic
PP-Growth (under development)

9. Mining interesting patterns from Streams

1. Frequent pattern mining

Basic
to be written

2. High utility pattern mining

Basic
HUPMS

10. Mining patterns from contiguous character sequences (E.g., DNA, Genome, and Game sequences)

10.1. Contiguous Frequent Patterns

Basic
PositionMining

11. Mining patterns from Graphs

11.1. Frequent sub-graph mining

Basic	topk
Gspan	TKG

11.2. Graph transactional coverage pattern mining

Basic
GTCP

12. Additional Features

12.1. Creation of synthetic databases

Database type
Transactional database
Temporal database
Utility database (coming soon)
spatio-transactional database (coming soon)
spatio-temporal database (coming soon)
fuzzy transactional database (coming soon)
fuzzy temporal database (coming soon)
Sequence database generator (coming soon)

12.2. Converting a dataframe into a specific database type

Approaches
Dense dataframe to databases
Sparse dataframe to databases (coming soon)

12.3. Gathering the statistical details of a database

Approaches
Transactional database
Temporal database
Utility database (coming soon)

12.4. Convertors

Approaches
Subgraphs2FlatTransactions
CSV2Parquet
CSV2BitInteger
CSV2Integer

12.4. Generating Latex code for the experimental results

Approaches
Latex code (coming soon)

---

Real World Case Studies

1. Air pollution analytics

Go to Top