LPPGrowth

class PAMI.localPeriodicPattern.basic.LPPGrowth.LPPGrowth(iFile, maxPer, maxSoPer, minDur, sep='\t')[source]

Bases: _localPeriodicPatterns

Description:

Local Periodic Patterns, which are patterns (sets of events) that have a periodic behavior in some non predefined time-intervals. A pattern is said to be a local periodic pattern if it appears regularly and continuously in some time-intervals. The maxSoPer (maximal period of spillovers) measure allows detecting time-intervals of variable lengths where a pattern is continuously periodic, while the minDur (minimal duration) measure ensures that those time-intervals have a minimum duration.

Reference:

Fournier-Viger, P., Yang, P., Kiran, R. U., Ventura, S., Luna, J. M.. (2020). Mining Local Periodic Patterns in a Discrete Sequence. Information Sciences, Elsevier, to appear. [ppt] DOI: 10.1016/j.ins.2020.09.044

Parameters:

iFile – str : Name of the Input file to mine complete set of local periodic pattern’s
oFile – str : Name of the output file to store complete set of local periodic patterns
minDur – str: Minimal duration in seconds between consecutive periods of time-intervals where a pattern is continuously periodic.
maxPer – float: Controls the maximum number of transactions in which any two items within a pattern can reappear.
maxSoPer – float: Controls the maximum number of time periods between consecutive periods of time-intervals where a pattern is continuously periodic.
sep – str : This variable is used to distinguish items from one another in a transaction. The default seperator is tab space. However, the users can override their default separator.

Attributes:

iFilestr: Input file name or path of the input file
oFilestr: Output file name or path of the output file
maxPerfloat: User defined maxPer value.
maxSoPerfloat: User defined maxSoPer value.
minDurfloat: User defined minDur value.
tsMinint / date: First time stamp of input data.
tsMaxint / date: Last time stamp of input data.
startTimefloat: Time when start of execution the algorithm.
endTimefloat: Time when end of execution the algorithm.
finalPatternsdict: To store local periodic patterns and its PTL.
tsListdict: To store items and its time stamp as bit vector.
rootTree: It is root node of transaction tree of whole input data.
PTLdict: Storing the item and its PTL.
itemslist: Storing local periodic item list.
sep: str: separator used to distinguish items from each other. The default separator is tab space.

Methods:

findSeparator(line): Find the separator of the line which split strings.
creteLPPlist(): Create the local periodic patterns list from input data.
createTSList(): Create the tsList as bit vector from input data.
generateLPP(): Generate 1 length local periodic pattens by tsList and execute depth first search.
createLPPTree(): Create LPPTree of local periodic item from input data.
patternGrowth(tree, prefix, prefixPFList): Execute pattern growth algorithm. It is important function in this program.
calculatePTL(tsList): Calculate PTL from input tsList as integer list.
calculatePTLbit(tsList): Calculate PTL from input tsList as bit vector.
mine(): Mining process will start from here.
getMemoryUSS(): Total amount of USS memory consumed by the mining process will be retrieved from this function.
getMemoryRSS(): Total amount of RSS memory consumed by the mining process will be retrieved from this function.
getRuntime(): Total amount of runtime taken by the mining process will be retrieved from this function.
getLocalPeriodicPatterns(): return local periodic patterns and its PTL
save(oFile): Complete set of local periodic patterns will be loaded in to an output file.
getPatternsAsDataFrame(): Complete set of local periodic patterns will be loaded in to a dataframe.

Executing the code on terminal:

Format:

(.venv) $ python3 LPPMGrowth.py <inputFile> <outputFile> <maxPer> <minSoPer> <minDur>

Example Usage:

(.venv) $ python3 LPPMGrowth.py sampleDB.txt patterns.txt 0.3 0.4 0.5

Sample run of importing the code:

from PAMI.localPeriodicPattern.basic import LPPGrowth as alg

obj = alg.LPPGrowth(iFile, maxPer, maxSoPer, minDur)

obj.mine()

localPeriodicPatterns = obj.getPatterns()

print(f'Total number of local periodic patterns: {len(localPeriodicPatterns)}')

obj.save(oFile)

Df = obj.getPatternsAsDataFrame()

memUSS = obj.getMemoryUSS()

print(f'Total memory in USS: {memUSS}')

memRSS = obj.getMemoryRSS()

print(f'Total memory in RSS: {memRSS}')

runtime = obj.getRuntime()

print(f'Total execution time in seconds: {runtime})

Credits:

The complete program was written by So Nakamura under the supervision of Professor Rage Uday Kiran.

getMemoryRSS() → float[source]

Total amount of RSS memory consumed by the mining process will be retrieved from this function

Returns:: returning RSS memory consumed by the mining process
Return type:: float

getMemoryUSS() → float[source]

Total amount of USS memory consumed by the mining process will be retrieved from this function

Returns:: returning USS memory consumed by the mining process
Return type:: float

getPatterns() → Dict[source]

Function to send the set of local periodic patterns after completion of the mining process

Returns:: returning frequent patterns
Return type:: dict

getPatternsAsDataFrame() → DataFrame[source]

Storing final local periodic patterns in a dataframe

Returns:: returning local periodic patterns in a dataframe
Return type:: pd.DataFrame

getRuntime() → float[source]

Calculating the total amount of runtime taken by the mining process

Returns:: returning total amount of runtime taken by the mining process
Return type:: float

mine() → None[source]: Mining process start from here.

printResults() → None[source]: This function is used to print the results

save(outFile: str) → None[source]

Complete set of local periodic patterns will be loaded in to an output file

Parameters:: outFile (csv file) – name of the output file
Returns:: None

startMine() → None[source]: Mining process start from here.

class PAMI.localPeriodicPattern.basic.LPPGrowth.Node[source]

Bases: object

A class used to represent the node of localPeriodicPatternTree

Attributes:

itemint: storing item of a node
parentnode: To maintain the parent of every node
childlist: To maintain the children of node
nodeLinknode: To maintain the next node of node
tidListset: To maintain timestamps of node

Methods:

getChild(itemName): storing the children to their respective parent nodes

getChild(item: int) → Node[source]

This function is used to get child node from the parent node

Parameters:: item (int) – item of the parent node
Returns:: if node have node of item, then return it. if node don’t have return []
Return type:: Node

class PAMI.localPeriodicPattern.basic.LPPGrowth.Tree[source]

Bases: object

A class used to represent the frequentPatternGrowth tree structure

Attributes:

rootnode: Represents the root node of the tree
nodeLinksdictionary: storing last node of each item
firstNodeLinkdictionary: storing first node of each item

Methods:

addTransaction(transaction,timeStamp): creating transaction as a branch in frequentPatternTree
fixNodeLinks(itemName, newNode): add newNode link after last node of item
deleteNode(itemName): delete all node of item
createPrefixTree(path,timeStampList): create prefix tree by path

addTransaction(transaction: List[int], tid: int) → None[source]

add transaction into tree

Parameters:

transaction (list) – it represents the one transaction in database
tid (list or int) – represents the timestamp of transaction

Returns:

None

createPrefixTree(path: List[int], tidList: List[int]) → None[source]

create prefix tree by path

Parameters:

path (list) – it represents path to root from prefix node
tidList (list) – it represents tid of each item

Returns:

None

deleteNode(item: int) → None[source]

delete the node from tree

Parameters:: item (str) – it represents the item name of node
Returns:: None

fixNodeLinks(item: int, newNode: Node) → None[source]

fix node link

Parameters:

item (string) – it represents item name of newNode
newNode (Node) – it represents node which is added

Returns:

None