# Fuzzy Periodic Frequent Pattern Miner is desired to find all fuzzy periodic frequent patterns which is
# on-trivial and challenging problem to its huge search space.we are using efficient pruning
# techniques to reduce the search space.
#
# Sample run of importing the code:
# ----------------------------------------
#
# from PAMI.fuzzyPeriodicFrequentPattern.basic import FPFPMiner as alg
#
# obj =alg.FPFPMiner("input.txt",2,3)
#
# obj.mine()
#
# periodicFrequentPatterns = obj.getPatterns()
#
# print("Total number of Fuzzy Periodic Frequent Patterns:", len(periodicFrequentPatterns))
#
# obj.save("output.txt")
#
# memUSS = obj.getMemoryUSS()
#
# print("Total Memory in USS:", memUSS)
#
# memRSS = obj.getMemoryRSS()
#
# print("Total Memory in RSS", memRSS)
#
# run = obj.getRuntime()
#
# print("Total ExecutionTime in seconds:", run)
#
__copyright__ = """
Copyright (C) 2021 Rage Uday Kiran
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <https://www.gnu.org/licenses/>.
"""
from PAMI.fuzzyPeriodicFrequentPattern.basic import abstract as _ab
from typing import List, Dict, Tuple, Set, Union, Any, Generator
from deprecated import deprecated
class _FFList:
"""
A class represent a Fuzzy List of an element
:Attributes:
item : int
the item name
sumLUtil : float
the sum of utilities of a fuzzy item in database
sumRUtil : float
the sum of resting values of a fuzzy item in database
elements : list
list of elements contain tid,Utility and resting values of element in each transaction
maxPeriod : int
it represents the max period of a item
:Methods:
addElement(element)
Method to add an element to this fuzzy list and update the sums at the same time.
printElement(e)
Method to print elements
"""
def __init__(self, itemName: str) -> None:
self.item = itemName
self.sumLUtil = 0.0
self.sumRUtil = 0.0
self.elements = []
self.maxPeriod = 0
def addElement(self, element) -> None:
"""
A Method that add a new element to FFList
:param element: an element to be added to FFList
:type element: Element
:return: None
"""
self.sumLUtil += element.lUtils
self.sumRUtil += element.rUtils
self.elements.append(element)
self.maxPeriod = max(self.maxPeriod, element.period)
def printElement(self) -> None:
"""
A Method to Print elements in the FFList
:return: None
"""
for ele in self.elements:
print(ele.tid, ele.lUtils, ele.rUtils, ele.period)
class _Element:
"""
A class represents an Element of a fuzzy list
:Attributes:
tid : int
keep tact of transaction id
lUtils: float
the utility of a fuzzy item in the transaction
rUtils : float
the resting value of a fuzzy item in the transaction
period: int
represent the period of the element
"""
def __init__(self, tid: int, iUtil: float, rUtil: float, period: int) -> None:
self.tid = tid
self.lUtils = iUtil
self.rUtils = rUtil
self.period = period
class _Pair:
"""
A class to store item name and quantity together.
"""
def __init__(self) -> None:
self.item = 0
self.quantity = 0
[docs]
class FPFPMiner(_ab._fuzzyPeriodicFrequentPatterns):
"""
:Description: Fuzzy Periodic Frequent Pattern Miner is desired to find all fuzzy periodic frequent patterns which is
on-trivial and challenging problem to its huge search space.we are using efficient pruning
techniques to reduce the search space.
:Reference: R. U. Kiran et al., "Discovering Fuzzy Periodic-Frequent Patterns in Quantitative Temporal Databases,"
2020 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), Glasgow, UK, 2020, pp.
1-8, doi: 10.1109/FUZZ48607.2020.9177579.
:param iFile: str :
Name of the Input file to mine complete set of frequent patterns
:param oFile: str :
Name of the output file to store complete set of frequent patterns
:param minSup: int or float or str :
The user can specify minSup either in count or proportion of database size. If the program detects the data type of minSup is integer, then it treats minSup is expressed in count. Otherwise, it will be treated as float.
:param maxPer: float :
The user can specify maxPer in count or proportion of database size. If the program detects the data type of maxPer is integer, then it treats maxPer is expressed in count.
:param 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:
iFile : file
Name of the input file to mine complete set of fuzzy spatial frequent patterns
oFile : file
Name of the oFile file to store complete set of fuzzy spatial frequent patterns
minSup : float
The user given support
period: int
periodicity of an element
memoryRSS : float
To store the total amount of RSS memory consumed by the program
startTime:float
To record the start time of the mining process
endTime:float
To record the completion time of the mining process
itemsCnt: int
To record the number of fuzzy spatial itemSets generated
mapItemsLowSum: map
To keep track of low region values of items
mapItemsMidSum: map
To keep track of middle region values of items
mapItemsHighSum: map
To keep track of high region values of items
mapItemSum: map
To keep track of sum of Fuzzy Values of items
mapItemRegions: map
To Keep track of fuzzy regions of item
jointCnt: int
To keep track of the number of FFI-list that was constructed
BufferSize: int
represent the size of Buffer
itemBuffer list
to keep track of items in buffer
maxTID: int
represent the maximum tid of the database
lastTIDs: map
represent the last tid of fuzzy items
itemsToRegion: map
represent items with respective regions
:Methods:
mine()
Mining process will start from here
getPatterns()
Complete set of patterns will be retrieved with this function
save(oFile)
Complete set of frequent patterns will be loaded in to a output file
getPatternsAsDataFrame()
Complete set of frequent patterns will be loaded in to a dataframe
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
convert(value):
To convert the given user specified value
FSFIMining( prefix, prefixLen, fsFim, minSup)
Method generate FFI from prefix
construct(px, py)
A function to construct Fuzzy itemSet from 2 fuzzy itemSets
findElementWithTID(UList, tid)
To find element with same tid as given
WriteOut(prefix, prefixLen, item, sumIUtil,period)
To Store the patten
**Executing the code on terminal :**
----------------------------------------
.. code-block:: console
Format:
(.venv) $ python3 FPFPMiner.py <inputFile> <outputFile> <minSup> <maxPer> <sep>
Example Usage:
(.venv) $ python3 FPFPMiner.py sampleTDB.txt output.txt 2 3
.. note:: minSup will be considered in percentage of database transactions
**Sample run of importing the code:**
--------------------------------------
from PAMI.fuzzyPeriodicFrequentPattern.basic import FPFPMiner as alg
obj =alg.FPFPMiner("input.txt",2,3)
obj.mine()
periodicFrequentPatterns = obj.getPatterns()
print("Total number of Fuzzy Periodic Frequent Patterns:", len(periodicFrequentPatterns))
obj.save("output.txt")
memUSS = obj.getMemoryUSS()
print("Total Memory in USS:", memUSS)
memRSS = obj.getMemoryRSS()
print("Total Memory in RSS", memRSS)
run = obj.getRuntime()
print("Total ExecutionTime in seconds:", run)
**Credits:**
--------------
The complete program was written by Sai Chitra.B under the supervision of Professor Rage Uday Kiran.
"""
_startTime = float()
_endTime = float()
_minSup = float()
_maxPer = float()
_finalPatterns = {}
_iFile = " "
_oFile = " "
_memoryUSS = float()
_memoryRSS = float()
_sep = " "
_Database = []
_transactions = []
_fuzzyValues = []
_ts = []
def __init__(self, iFile: Union[str, _ab._pd.DataFrame], minSup: Union[int, float], period: Union[int, float], sep: str="\t") -> None:
super().__init__(iFile, minSup, period, sep)
self._oFile = ""
self._BufferSize = 200
self._itemSetBuffer = []
self._mapItemSum = {}
self._finalPatterns = {}
self._joinsCnt = 0
self._itemsCnt = 0
self._startTime = float()
self._endTime = float()
self._memoryUSS = float()
self._memoryRSS = float()
self._dbLen = 0
def _compareItems(self, o1, o2) -> int:
"""
A Function that sort all FFI-list in ascending order of Support
:param o1: First FFI-list
:type o1: _FFList
:param o2: Second FFI-list
:type o1: _FFList
:return: Comparision Value
:rtype: int
"""
compare = self._mapItemSum[o1.item] - self._mapItemSum[o2.item]
if compare == 0:
return int(o1.item) - int(o2.item)
else:
return compare
def _convert(self, value) -> float:
"""
To convert the given user specified value
:param value: user specified value
:type value: int or float or str
:return: converted value
:rtype: float
"""
if type(value) is int:
value = int(value)
if type(value) is float:
value = (self._dbLen * value)
if type(value) is str:
if '.' in value:
value = (self._dbLen * value)
else:
value = int(value)
return value
def _creatingItemSets(self) -> None:
"""
Storing the complete transactions of the database/input file in a database variable
:return: None
"""
data, self._transactions, self._fuzzyValues, ts = [], [], [], []
if isinstance(self._iFile, _ab._pd.DataFrame):
if self._iFile.empty:
print("its empty..")
i = self._iFile.columns.values.tolist()
if 'TS' in i:
self._ts = self._iFile['TS'].tolist()
if 'Transactions' in i:
self._transactions = self._iFile['Transactions'].tolist()
if 'fuzzyValues' in i:
self._fuzzyValues = self._iFile['fuzzyValues'].tolist()
if isinstance(self._iFile, str):
if _ab._validators.url(self._iFile):
data = _ab._urlopen(self._iFile)
count = 0
for line in data:
line = line.decode("utf-8")
line = line.split("\n")[0]
parts = line.split(":")
parts[0] = parts[0].strip()
parts[1] = parts[1].strip()
items = parts[0].split(self._sep)
quantities = parts[1].split(self._sep)
self._ts.append(int(items[0]))
self._transactions.append([x for x in items[1:]])
self._fuzzyValues.append([float(x) for x in quantities])
count += 1
else:
try:
with open(self._iFile, 'r', encoding='utf-8') as f:
count = 0
for line in f:
line = line.split("\n")[0]
parts = line.split(":")
parts[0] = parts[0].strip()
parts[1] = parts[1].strip()
items = parts[0].split(self._sep)
quantities = parts[1].split(self._sep)
self._ts.append(int(items[0]))
self._transactions.append([x for x in items[1:]])
self._fuzzyValues.append([float(x) for x in quantities])
count += 1
except IOError:
print("File Not Found")
quit()
[docs]
@deprecated("It is recommended to use 'mine()' instead of 'mine()' for mining process. Starting from January 2025, 'mine()' will be completely terminated.")
def startMine(self) -> None:
"""
Fuzzy periodic Frequent pattern mining process will start from here
"""
self.mine()
[docs]
def mine(self) -> None:
"""
Fuzzy periodic Frequent pattern mining process will start from here
"""
maxTID = 0
lastTIDs = {}
self._startTime = _ab._time.time()
self._creatingItemSets()
self._finalPatterns = {}
tid = int()
for line in range(len(self._transactions)):
tid = int(self._ts[line])
self._dbLen += 1
items = self._transactions[line]
quantities = self._fuzzyValues[line]
if tid < maxTID:
maxTID = tid
for i in range(0, len(items)):
item = items[i]
if item in self._mapItemSum:
self._mapItemSum[item] += quantities[i]
else:
self._mapItemSum[item] = quantities[i]
listOfFFIList = []
mapItemsToFFLIST = {}
# self._minSup = self._convert(self._minSup)
self._minSup = float(self._minSup)
self._maxPer = self._convert(self._maxPer)
for item1 in self._mapItemSum.keys():
item = item1
if self._mapItemSum[item] >= self._minSup:
fUList = _FFList(item)
k = tuple([item])
mapItemsToFFLIST[k] = fUList
listOfFFIList.append(fUList)
lastTIDs[item] = tid
listOfFFIList.sort(key=_ab._functools.cmp_to_key(self._compareItems))
for line in range(len(self._transactions)):
tid = int(self._ts[line])
items = self._transactions[line]
quantities = self._fuzzyValues[line]
revisedTransaction = []
for i in range(0, len(items)):
pair = _Pair()
pair.item = items[i]
item = pair.item
pair.quantity = quantities[i]
if self._mapItemSum[item] >= self._minSup:
if pair.quantity > 0:
revisedTransaction.append(pair)
revisedTransaction.sort(key=_ab._functools.cmp_to_key(self._compareItems))
for i in range(len(revisedTransaction) - 1, -1, -1):
pair = revisedTransaction[i]
remainUtil = 0
for j in range(len(revisedTransaction) - 1, i - 1, -1):
remainUtil += revisedTransaction[j].quantity
if pair.quantity > remainUtil:
remainingUtility = pair.quantity
else:
remainingUtility = remainUtil
if mapItemsToFFLIST.get(tuple([pair.item])) is not None:
FFListOfItem = mapItemsToFFLIST[tuple([pair.item])]
if len(FFListOfItem.elements) == 0:
element = _Element(tid, pair.quantity, remainingUtility, 0)
else:
if lastTIDs[pair.item] == tid:
element = _Element(tid, pair.quantity, remainingUtility, maxTID - tid)
else:
lastTid = FFListOfItem.elements[-1].tid
curPer = tid - lastTid
element = _Element(tid, pair.quantity, remainingUtility, curPer)
FFListOfItem.addElement(element)
self._FPFPMining(self._itemSetBuffer, 0, listOfFFIList)
self._endTime = _ab._time.time()
process = _ab._psutil.Process(_ab._os.getpid())
self._memoryUSS = float()
self._memoryRSS = float()
self._memoryUSS = process.memory_full_info().uss
self._memoryRSS = process.memory_info().rss
def _FPFPMining(self, prefix, prefixLen, fsFim):
"""
Generates FPFP from prefix
:param prefix: the prefix patterns of FPFP
:type prefix: len
:param prefixLen: the length of prefix
:type prefixLen: int
:param fsFim: the Fuzzy list of prefix itemSets
:type fsFim: list
"""
for i in range(0, len(fsFim)):
X = fsFim[i]
if X.sumLUtil >= self._minSup and X.maxPeriod <= self._maxPer:
self._WriteOut(prefix, prefixLen, X.item, X.sumLUtil, X.maxPeriod)
if X.sumRUtil >= self._minSup:
exULs = []
for j in range(i + 1, len(fsFim)):
Y = fsFim[j]
exULs.append(self._construct(X, Y))
self._joinsCnt += 1
self._itemSetBuffer.insert(prefixLen, X.item)
self._FPFPMining(self._itemSetBuffer, prefixLen + 1, exULs)
[docs]
def getMemoryUSS(self) -> float:
"""
Total amount of USS memory consumed by the mining process will be retrieved from this function
:return: returning USS memory consumed by the mining process
:rtype: float
"""
return self._memoryUSS
[docs]
def getRuntime(self) -> float:
"""
Calculating the total amount of runtime taken by the mining process
:return: returning total amount of runtime taken by the mining process
:rtype: float
"""
return self._endTime - self._startTime
def _construct(self, px: _FFList, py: _FFList) -> _FFList:
"""
A function to construct a new Fuzzy item set from 2 fuzzy itemSets
:param px:the item set px
:type px:FFI-List
:param py:item set py
:type py:FFI-List
:return :the item set of pxy(px and py)
:rtype :FFI-List
"""
pxyUL = _FFList(py.item)
prev = 0
for ex in px.elements:
ey = self._findElementWithTID(py, ex.tid)
if ey is None:
continue
eXY = _Element(ex.tid, min([ex.lUtils, ey.lUtils], key=lambda x: float(x)), ey.rUtils, ex.tid - prev)
pxyUL.addElement(eXY)
prev = ex.tid
return pxyUL
def _findElementWithTID(self, UList, tid) -> _Element:
"""
To find element with same tid as given
:param UList: fuzzy list
:type UList: FFI-List
:param tid:transaction id
:type tid: int
:return: element eith tid as given
:rtype: element if exist or None
"""
List_ = UList.elements
first = 0
last = len(List_) - 1
while first <= last:
mid = (first + last) >> 1
if List_[mid].tid < tid:
first = mid + 1
elif List_[mid].tid > tid:
last = mid - 1
else:
return List_[mid]
return None
def _WriteOut(self, prefix: List[int], prefixLen: int, item: int, sumLUtil: float, period: int) -> None:
"""
To Store the patten
:param prefix: prefix of itemSet
:type prefix: list
:param prefixLen: length of prefix
:type prefixLen: int
:param item: the last item
:type item: int
:param sumLUtil: sum of utility of itemSet
:type sumLUtil: float
:param period: represent the period of itemSet
:type period: int
:return: None
"""
self._itemsCnt += 1
res = ""
for i in range(0, prefixLen):
res += str(prefix[i]) + "\t"
res += str(item)
#res1 = str(sumLUtil) + " : " + str(period)
self._finalPatterns[res] = [sumLUtil, period]
[docs]
def getPatternsAsDataFrame(self) -> _ab._pd.DataFrame:
"""
Storing final frequent patterns in a dataframe
:return: returning frequent patterns in a dataframe
:rtype: pd.DataFrame
"""
dataFrame = {}
data = []
for a, b in self._finalPatterns.items():
data.append([a.replace('\t', ' '), b[0], b[1]])
dataFrame = _ab._pd.DataFrame(data, columns=['Patterns', 'Support', 'Periodicity'])
return dataFrame
[docs]
def getPatterns(self) -> Dict[str, str]:
"""
Function to send the set of frequent patterns after completion of the mining process
:return: returning frequent patterns
:rtype: dict
"""
return self._finalPatterns
[docs]
def save(self, outFile: str) -> None:
"""
Complete set of frequent patterns will be loaded in to an output file
:param outFile: name of the output file
:type outFile: csv file
:return: None
"""
self._oFile = outFile
writer = open(self._oFile, 'w+')
for x, y in self._finalPatterns.items():
patternsAndSupport = x.strip() + ":" + str(y[0]) + ":" + str(y[1])
writer.write("%s \n" % patternsAndSupport)
[docs]
def printResults(self) -> None:
"""
This function is used to print the results
"""
print("Total number of Fuzzy Periodic-Frequent Patterns:", len(self.getPatterns()))
print("Total Memory in USS:", self.getMemoryUSS())
print("Total Memory in RSS", self.getMemoryRSS())
print("Total ExecutionTime in seconds:", self.getRuntime())
if __name__ == "__main__":
_ap = str()
if len(_ab._sys.argv) == 5 or len(_ab._sys.argv) == 6:
if len(_ab._sys.argv) == 6: # to include a user specified separator
_ap = FPFPMiner(_ab._sys.argv[1], _ab._sys.argv[3], _ab._sys.argv[4], _ab._sys.argv[5])
if len(_ab._sys.argv) == 5: # to consider "\t" as a separator
_ap = FPFPMiner(_ab._sys.argv[1], _ab._sys.argv[3], _ab._sys.argv[4])
_ap.mine()
_ap.mine()
print("Total number of Fuzzy Periodic-Frequent Patterns:", len(_ap.getPatterns()))
_ap.save(_ab._sys.argv[2])
print("Total Memory in USS:", _ap.getMemoryUSS())
print("Total Memory in RSS", _ap.getMemoryRSS())
print("Total ExecutionTime in seconds:", _ap.getRuntime())
else:
_ap = FPFPMiner('sample.txt', 1, 10, ' ')
_ap.mine()
_ap.mine()
print("Total number of Fuzzy Periodic-Frequent Patterns:", len(_ap.getPatterns()))
_ap.save('output.txt')
print("Total Memory in USS:", _ap.getMemoryUSS())
print("Total Memory in RSS", _ap.getMemoryRSS())
print("Total ExecutionTime in seconds:", _ap.getRuntime())
print("Error! The number of input parameters do not match the total number of parameters provided")