def test01():
print("Test 01")
m = PModelBandit(["r1", "r2"])
print(m.head(10))
print("")
df = DataFrame([[0, 1, 1, 1, 1]], columns=PModelBandit.getColumns())
df.set_index(PModelBandit.COL_METHOD_ID, inplace=True)
df.__class__ = PModelBandit
def test01():
print("Simulation: ML BanditTS")
jobID: str = "BanditTS"
selector = RouletteWheelSelector({RouletteWheelSelector.ARG_EXPONENT: 1})
rIDs, rDescs = InputRecomSTDefinition.exportPairOfRecomIdsAndRecomDescrs()
pDescr: Portfolio1AggrDescription = Portfolio1AggrDescription(
"BanditTS" + jobID, rIDs, rDescs,
InputAggrDefinition.exportADescBanditTS(selector))
batchID: str = "ml1mDiv90Ulinear0109R1"
dataset: DatasetML = DatasetML.readDatasets()
behaviourFile: str = BehavioursML.getFile(BehavioursML.BHVR_LINEAR0109)
behavioursDF: DataFrame = BehavioursML.readFromFileMl1m(behaviourFile)
model: DataFrame = PModelBandit(pDescr.getRecommendersIDs())
# simulation of portfolio
simulator: Simulator = Simulator(batchID, SimulationML, argsSimulationDict,
dataset, behavioursDF)
simulator.simulate([pDescr], [model], [EvalToolBanditTS({})],
[HistoryHierDF(pDescr.getPortfolioID())])
def run(self, batchID: str, jobID: str):
divisionDatasetPercentualSize: int
uBehaviour: str
repetition: int
divisionDatasetPercentualSize, uBehaviour, repetition = InputABatchDefinition(
).getBatchParameters(self.datasetID)[batchID]
selector: ADHondtSelector = self.getParameters()[jobID]
rIDs, rDescs = InputRecomSTDefinition.exportPairOfRecomIdsAndRecomDescrs(
)
pDescr: Portfolio1AggrDescription = Portfolio1AggrDescription(
self.getBatchName() + jobID, rIDs, rDescs,
InputAggrDefinition.exportADescBanditTS(selector))
eTool: AEvalTool = EvalToolBanditTS({})
model: DataFrame = PModelBandit(pDescr.getRecommendersIDs())
simulator: Simulator = InputSimulatorDefinition(
).exportSimulatorSlantour(batchID, divisionDatasetPercentualSize,
uBehaviour, repetition)
simulator.simulate([pDescr], [model], [eTool],
[HistoryHierDF(pDescr.getPortfolioID())])
def run(self, batchID: str, jobID: str):
divisionDatasetPercentualSize: int
uBehaviour: str
repetition: int
divisionDatasetPercentualSize, uBehaviour, repetition = \
InputABatchDefinition().getBatchParameters(self.datasetID)[batchID]
recommenderTheMPopID: str = "TheMostPopular"
pRTheMPopDescr: RecommenderDescription = RecommenderDescription(
RecommenderTheMostPopular, {})
recommenderRPID: str = "RepeatedPurchase"
pRecRPDescr: RecommenderDescription = RecommenderDescription(
RecommenderRepeatedPurchase, {})
selector: ADHondtSelector = self.getParameters()[jobID]
aDescDHont: AggregationDescription = InputAggrDefinition.exportADescDHondtDirectOptimizeThompsonSampling(
selector)
aDescDHont: AggregationDescription = InputAggrDefinition.exportADescBanditTS(
selector)
#aDescDHont:AggregationDescription = InputAggrDefinition.exportADescFAI()
rIDs: List[str]
rDescs: List[AggregationDescription]
rIDs, rDescs = InputRecomRRDefinition.exportPairOfRecomIdsAndRecomDescrs(
)
#rIDs = [recommenderTheMPopID]
#rDescs = [pRTheMPopDescr]
p1AggrDescrID: str = "p1AggrDescrID"
p1AggrDescr: Portfolio1AggrDescription = Portfolio1AggrDescription(
p1AggrDescrID, rIDs, rDescs, aDescDHont)
#pProbTool:APenalization = PenalizationToolDefinition.exportProbPenaltyToolOLin0802HLin1002(
# InputSimulatorDefinition().numberOfAggrItems)
pProbTool: APenalization = PenalizationToolDefinition.exportPenaltyToolOStat08HLin1002(
InputSimulatorDefinition().numberOfAggrItems)
aHierDescr: AggregationDescription = AggregationDescription(
AggrD21, {AggrD21.ARG_RATING_THRESHOLD_FOR_NEG: 0.0})
pHierDescr: PortfolioHierDescription = PortfolioHierDescription(
"pHierDescr", recommenderRPID, pRecRPDescr, p1AggrDescrID,
p1AggrDescr, aHierDescr, pProbTool)
eTool: AEvalTool = EvalToolBanditTS({})
#eTool:AEvalTool = EToolDoNothing({})
#model:DataFrame = PModelDHont(p1AggrDescr.getRecommendersIDs())
model: DataFrame = PModelBandit(p1AggrDescr.getRecommendersIDs())
simulator: Simulator = InputSimulatorDefinition(
).exportSimulatorRetailRocket(batchID, divisionDatasetPercentualSize,
uBehaviour, repetition)
simulator.simulate([pHierDescr], [model], [eTool],
[HistoryHierDF(p1AggrDescr.getPortfolioID())])
def getBanditTS():
taskID:str = "Web" + "BanditTS" + "Roulette1"
selector = RouletteWheelSelector({RouletteWheelSelector.ARG_EXPONENT: 1})
rIDs, rDescs = InputRecomSTDefinition.exportPairOfRecomIdsAndRecomDescrs()
pDescr:Portfolio1AggrDescription = Portfolio1AggrDescription(
taskID, rIDs, rDescs, InputAggrDefinition.exportADescBanditTS(selector))
dataset:DatasetST = DatasetST.readDatasets()
history:AHistory = HistoryHierDF(taskID)
port:APortfolio = pDescr.exportPortfolio(taskID, history)
port.train(history, dataset)
model:DataFrame = PModelBandit(pDescr.getRecommendersIDs())
evalTool:AEvalTool = EvalToolBanditTS({})
return (taskID, port, model, evalTool, history)
scores: List[float] = [x[1] for x in result]
resultSer: Series = Series(scores, index=itemsIDs)
finalScores = normalize(np.expand_dims(resultSer.values, axis=0))[0, :]
resultNorm: List[tuple] = zip(resultSer.index, finalScores.tolist())
return resultNorm
if __name__ == "__main__":
os.chdir("..")
rItemIdsWitRespons: List[tuple] = [(11, 'metoda1'), (21, 'metoda2')]
modelData: List[tuple] = [['metoda1', 5, 10, 1,
1], ['metoda2', 1, 10, 1, 1],
['metoda3', 2, 20, 1, 1]]
modelDF: DataFrame = pd.DataFrame(
modelData, columns=["methodID", "r", "n", "alpha0", "beta0"])
modelDF.set_index("methodID", inplace=True)
modelDF: DataFrame = PModelBandit(['metoda1', 'metoda2', 'metoda3'])
print("Model:")
print(modelDF)
print("")
result: List[tuple] = countAggrBanditsResponsibility(
rItemIdsWitRespons, modelDF)
print("Result:")
print(result)