def __init__(self,tuneID,params,data,queue):
multiprocessing.Process.__init__(self)
self.tuneID=tuneID
self.model=CascadingModel(**params)
#if transferLearning:
self.data=data
self.queue=queue
self.params=params
def testBestReRank(k,bestModel=None):
if bestModel is None:
model=CascadingModel(tasktype=modeltype,digtype=datatype)
else:
model=bestModel
mymetric=model.mymetric
mymetric.verbose=0
taskids=data.taskids[:data.testPoint]
Y_label=data.testLabel[:data.testPoint]
print("tuning re-rank weight\n")
best_param={3:0,5:0,10:0}
rw=[w/10 for w in range(0,11)]
rw.reverse()
model.topK=k
maxAcc=[0,0]
model.loadConf()
model.loadModel()
Y_predict2=model.predict(data.testX,taskids)
for w in rw:
acc=mymetric.topKPDIGUsers(Y_predict2,Y_label,taskids,k,w)
acc=np.mean(acc)
if acc>maxAcc[0]:
maxAcc=[acc,w]
#print(data.tasktype,"top %d"%k,acc,"weight=%f"%(w/10))
best_param[k]=maxAcc[1]
print("\n",data.tasktype,"top %d"%k,maxAcc[0],"weight=%f"%maxAcc[1])
print()
def testCascadingModel():
model=CascadingModel(tasktype=modeltype)
taskids=data.taskids[:data.testPoint]
Y_label=data.testLabel[:data.testPoint]
Y_sublabel=data.submitLabelClassification[:data.testPoint]
Y_reglabel=data.registerLabelClassification[:data.testPoint]
mymetric=model.mymetric
mymetric.callall=False
print()
print("\n meta-learning model top k acc")
for k in (3,5,10):
model.topK=k
model.loadConf()
model.loadModel()
Y_predict2=model.predict(data.testX,taskids)
#metrics
acc=mymetric.topKPossibleUsers(Y_predict2,Y_label,k)
acc=np.mean(acc)
print(data.tasktype,"top %d"%k,acc)
acc=mymetric.topKRUsers(Y_predict2,Y_label,Y_reglabel,k,)
acc=np.mean(acc)
print(data.tasktype,"top %d"%k,acc)
acc=mymetric.topKSUsers(Y_predict2,Y_label,Y_sublabel,k,)
acc=np.mean(acc)
print(data.tasktype,"top %d"%k,acc)
print()
class TuneTask(multiprocessing.Process):
maxProcessNum=32
def __init__(self,tuneID,params,data,queue):
multiprocessing.Process.__init__(self)
self.tuneID=tuneID
self.model=CascadingModel(**params)
#if transferLearning:
self.data=data
self.queue=queue
self.params=params
def run(self):
bestScore=0
topDig=[w/10 for w in range(10,11)]
topDig.reverse()
tD=0
#print("begin,%d"%self.tuneID)
mymetric=self.model.mymetric
taskids=self.data.taskids[self.data.validatePoint:]
X=self.data.trainX
Y_label=self.data.trainLabel
for self.model.topDig in topDig:
#print("predicting")
Y=self.model.predict(X,taskids)
Y=mymetric.topKPossibleUsers(Y,Y_label,self.model.topK)
acc=np.mean(Y)
#acc=self.model.score(self.data)
#print(acc,"topDig=%f"%self.model.topDig)
if acc>bestScore:
tD=self.model.topDig
bestScore=acc
self.model.topDig=tD
self.params["topDig"]=self.model.topDig
#print("finished")
#self.cond.acquire()
self.queue.put([self.tuneID,self.model,bestScore,self.params])