本文整理汇总了Python中pybrain.rl.experiments.Experiment._oneInteraction方法的典型用法代码示例。如果您正苦于以下问题：Python Experiment._oneInteraction方法的具体用法？Python Experiment._oneInteraction怎么用？Python Experiment._oneInteraction使用的例子？那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类pybrain.rl.experiments.Experiment的用法示例。
在下文中一共展示了Experiment._oneInteraction方法的1个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: BoxSearchRunner
# 需要导入模块: from pybrain.rl.experiments import Experiment [as 别名]
# 或者: from pybrain.rl.experiments.Experiment import _oneInteraction [as 别名]
class BoxSearchRunner():
def __init__(self, mode):
self.mode = mode
cu.mem('Reinforcement Learning Started')
self.environment = BoxSearchEnvironment(config.get(mode+'Database'), mode, config.get(mode+'GroundTruth'))
self.controller = QNetwork()
cu.mem('QNetwork controller created')
self.learner = None
self.agent = BoxSearchAgent(self.controller, self.learner)
self.task = BoxSearchTask(self.environment, config.get(mode+'GroundTruth'))
self.experiment = Experiment(self.task, self.agent)
def runEpoch(self, interactions, maxImgs):
img = 0
s = cu.tic()
while img < maxImgs:
k = 0
while not self.environment.episodeDone and k < interactions:
self.experiment._oneInteraction()
k += 1
self.agent.learn()
self.agent.reset()
self.environment.loadNextEpisode()
img += 1
s = cu.toc('Run epoch with ' + str(maxImgs) + ' episodes', s)
def run(self):
if self.mode == 'train':
self.agent.persistMemory = True
self.agent.startReplayMemory(len(self.environment.imageList), config.geti('trainInteractions'))
self.train()
elif self.mode == 'test':
self.agent.persistMemory = False
self.test()
def train(self):
networkFile = config.get('networkDir') + config.get('snapshotPrefix') + '_iter_' + config.get('trainingIterationsPerBatch') + '.caffemodel'
interactions = config.geti('trainInteractions')
minEpsilon = config.getf('minTrainingEpsilon')
epochSize = len(self.environment.imageList)/1
epsilon = 1.0
self.controller.setEpsilonGreedy(epsilon, self.environment.sampleAction)
epoch = 1
exEpochs = config.geti('explorationEpochs')
while epoch <= exEpochs:
s = cu.tic()
print 'Epoch',epoch,': Exploration (epsilon=1.0)'
self.runEpoch(interactions, len(self.environment.imageList))
self.task.flushStats()
self.doValidation(epoch)
s = cu.toc('Epoch done in ',s)
epoch += 1
self.learner = QLearning()
self.agent.learner = self.learner
egEpochs = config.geti('epsilonGreedyEpochs')
while epoch <= egEpochs + exEpochs:
s = cu.tic()
epsilon = epsilon - (1.0-minEpsilon)/float(egEpochs)
if epsilon < minEpsilon: epsilon = minEpsilon
self.controller.setEpsilonGreedy(epsilon, self.environment.sampleAction)
print 'Epoch',epoch ,'(epsilon-greedy:{:5.3f})'.format(epsilon)
self.runEpoch(interactions, epochSize)
self.task.flushStats()
self.doValidation(epoch)
s = cu.toc('Epoch done in ',s)
epoch += 1
maxEpochs = config.geti('exploitLearningEpochs') + exEpochs + egEpochs
while epoch <= maxEpochs:
s = cu.tic()
print 'Epoch',epoch,'(exploitation mode: epsilon={:5.3f})'.format(epsilon)
self.runEpoch(interactions, epochSize)
self.task.flushStats()
self.doValidation(epoch)
s = cu.toc('Epoch done in ',s)
shutil.copy(networkFile, networkFile + '.' + str(epoch))
epoch += 1
def test(self):
interactions = config.geti('testInteractions')
self.controller.setEpsilonGreedy(config.getf('testEpsilon'))
self.runEpoch(interactions, len(self.environment.imageList))
def doValidation(self, epoch):
if epoch % config.geti('validationEpochs') != 0:
return
auxRL = BoxSearchRunner('test')
auxRL.run()
indexType = config.get('evaluationIndexType')
category = config.get('category')
if indexType == 'pascal':
categories, catIndex = bse.get20Categories()
elif indexType == 'relations':
categories, catIndex = bse.getCategories()
elif indexType == 'finetunedRelations':
categories, catIndex = bse.getRelationCategories()
if category in categories:
catI = categories.index(category)
else:
catI = -1
#.........这里部分代码省略.........