本文整理汇总了Python中utilities.Utilities.getTrainingFile方法的典型用法代码示例。如果您正苦于以下问题：Python Utilities.getTrainingFile方法的具体用法？Python Utilities.getTrainingFile怎么用？Python Utilities.getTrainingFile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utilities.Utilities的用法示例。
在下文中一共展示了Utilities.getTrainingFile方法的4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: __init__
# 需要导入模块: from utilities import Utilities [as 别名]
# 或者: from utilities.Utilities import getTrainingFile [as 别名]
def __init__(self, currentTime, outputDataType, numberOfExperts):
self.currentTime = currentTime
self.numberOfExperts = numberOfExperts
self.inputTrainingSetFile = Utilities.getTrainingFile(currentTime, DocumentType.typeRaw, self.numberOfExperts)
self.inputTestSetFile = Utilities.getTestFile(
currentTime, DocumentType.typeRaw, self.numberOfExperts, bottom=True
)
self.outputTrainingSetFile = Utilities.getTrainingFile(currentTime, outputDataType, self.numberOfExperts)
self.outputTestSetFile = Utilities.getTestFile(currentTime, outputDataType, self.numberOfExperts, bottom=True)
Utilities.createDirectory(self.outputTrainingSetFile), Utilities.createDirectory(self.outputTestSetFile)示例2: generateRawDataForGivenSetOfUsers
# 需要导入模块: from utilities import Utilities [as 别名]
# 或者: from utilities.Utilities import getTrainingFile [as 别名]
def generateRawDataForGivenSetOfUsers(numberOfUsersList):
for numberOfUsers in numberOfUsersList:
currentTime = Settings.startTime
allExpertsTop, allExpertsBottom = (
ExpertUsers(number=numberOfUsers),
ExpertUsers(number=numberOfUsers, type=ExpertUsers.typeBottom),
)
allExpertsList = {}
for k, v in allExpertsTop.list.iteritems():
allExpertsList[k] = v
for k, v in allExpertsBottom.list.iteritems():
allExpertsList[k] = v
while currentTime <= Settings.endTime:
trainingFile = Utilities.getTrainingFile(currentTime, DocumentType.typeRaw, numberOfUsers)
testFile = Utilities.getTestFile(currentTime, DocumentType.typeRaw, numberOfUsers, bottom=True)
Utilities.createDirectory(trainingFile), Utilities.createDirectory(testFile)
print numberOfUsers, Settings.twitterUsersTweetsFolder + "%s.gz" % Utilities.getDataFile(currentTime)
for tweet in CreateTrainingAndTestSets.getTweetsFromExperts(
allExpertsList, Settings.twitterUsersTweetsFolder + "%s.gz" % Utilities.getDataFile(currentTime)
):
tweet["class"] = allExpertsList[tweet["user"]["id_str"]]["class"]
if tweet["user"]["id_str"] in allExpertsTop.list:
Utilities.writeAsJsonToFile(tweet, trainingFile)
else:
Utilities.writeAsJsonToFile(tweet, testFile)
currentTime += timedelta(days=1)示例3: generateDataSetStats125
# 需要导入模块: from utilities import Utilities [as 别名]
# 或者: from utilities.Utilities import getTrainingFile [as 别名]
def generateDataSetStats125():
currentDay = Settings.startTime
while currentDay<=Settings.endTime:
data = {'day': datetime.strftime(currentDay, Settings.twitter_api_time_format), 'train_classes': defaultdict(int), 'test_classes': defaultdict(int)}
inputTrainingSetFile = Utilities.getTrainingFile(currentDay, DocumentType.typeRuuslUnigram, Settings.numberOfExperts)
inputTestSetFile = Utilities.getTestFile(currentDay, DocumentType.typeRuuslUnigram, Settings.numberOfExperts, bottom=True)
for file, tweetType in [(inputTrainingSetFile, 'training'), (inputTestSetFile, 'test')]:
for tweet in Utilities.iterateTweetsFromFile(file):
if tweetType=='training': data['train_classes'][tweet['class']]+=1
else: data['test_classes'][tweet['class']]+=1
Utilities.writeAsJsonToFile(data, Settings.stats_for_dataset_125)
currentDay+=timedelta(days=1)示例4: generateRawDataForIntermediateSetOfUsers
# 需要导入模块: from utilities import Utilities [as 别名]
# 或者: from utilities.Utilities import getTrainingFile [as 别名]
def generateRawDataForIntermediateSetOfUsers():
currentTime = Settings.startTime
allExpertsTop = ExpertUsers(Settings.numberOfExperts)
allExpertsIntermediate = ExpertUsers(Settings.numberOfExpertsSecondSet)
allExpertsList = {}
for k, v in allExpertsTop.list.iteritems():
del allExpertsIntermediate.list[k]
for k, v in allExpertsIntermediate.list.iteritems():
allExpertsList[k] = v
while currentTime <= Settings.endTime:
for numberOfExperts in [Settings.numberOfExpertsSecondSet]:
trainingFile = Utilities.getTrainingFile(currentTime, DocumentType.typeRaw, numberOfExperts)
Utilities.createDirectory(trainingFile)
print numberOfExperts, Settings.twitterUsersTweetsFolder + "%s.gz" % Utilities.getDataFile(currentTime)
for tweet in CreateTrainingAndTestSets.getTweetsFromExperts(
allExpertsList, Settings.twitterUsersTweetsFolder + "%s.gz" % Utilities.getDataFile(currentTime)
):
tweet["class"] = allExpertsList[tweet["user"]["id_str"]]["class"]
Utilities.writeAsJsonToFile(tweet, trainingFile)
currentTime += timedelta(days=1)