Python normalArm 예제들

예제 #1

파일: test_experiment.py 프로젝트: v-i-s-h/banditPy

from algorithms import UCB
from models import MAB
from models import experiment

# Plot tools
import matplotlib.pyplot as plt

# Create some arms
# arm1 = bernoulliArm.bernoulliArm( 0.70, reward1 = 5.0, reward2 = -4.0 )
# arm2 = bernoulliArm.bernoulliArm( 0.35, reward1 = 3.0 )
# arm3 = bernoulliArm.bernoulliArm( 0.40, reward1 = 3.0 )
# arm4 = bernoulliArm.bernoulliArm( 0.90, reward1 = 1.5 )
# armModels = [ arm1, arm2, arm3, arm4 ]

armModels = [
                normalArm.normalArm( 3.0 ),
                normalArm.normalArm( 4.0, 1.0 ),
                normalArm.normalArm( 5.0, 1.0 )
            ]

bandit = MAB.MAB( armModels )

algo0 = epsilonGreedy.epsilonGreedy( 0.25, len(armModels) )
algo1 = UCB.UCB1( len(armModels) )

exp0 = experiment.experiment( bandit, algo0 )
exp1 = experiment.experiment( bandit, algo1 )

# run expriment
selArmIndex0, obtainedRewards0 = exp0.run( 1000 )
selArmIndex1, obtainedRewards1 = exp1.run( 1000 )

예제 #2

파일: test_MAB.py 프로젝트: v-i-s-h/banditPy

# Changelogs

################################################################################

# Import root path
import os.path
import sys
sys.path.append(os.path.abspath(os.path.join(os.path.dirname(__file__), "../../modules")))

# Import modules
from arms import bernoulliArm, normalArm
from models import MAB

# create some arms
bArm1 = bernoulliArm.bernoulliArm( 0.7, reward1 = 3.0, reward2 = 2.5 )
bArm2 = bernoulliArm.bernoulliArm( 0.85 )
nArm1 = normalArm.normalArm( 3.0, 2.0 )
nArm2 = normalArm.normalArm( 2.0, 5.0 )

banditArms = [ bArm1, bArm2, nArm1, nArm2 ]

# Create a bandit model using arm
bandit = MAB.MAB( banditArms )

# Pull, pull and pull!!
print bandit.pull( 1 )
print bandit.pull( 0 )
print bandit.pull( [ 1, 3] )
print bandit.pull( [] )
print bandit.tick()