def speak(alias):
sounds_dir = Utilities.voices_path()
save_dir = Utilities.absolute_path(sounds_dir, alias + '.mp3')
if os.path.exists(save_dir) is True:
# print save_dir
pygame.mixer.init(44100, -16, 2, 2048)
pygame.mixer.music.load(save_dir)
pygame.mixer.music.play()
time.sleep(1.05)
def __init__(self, logFilename=None, utilObj=None):
if (utilObj!=None):
self.util=utilObj
elif(logFilename!=None):
self.util = Utilities.Utility()
self.util.setupLogFileLoc(logFilename)
self.util.startTimeTrack()
def __init__(self,
vectordim=100,
wordFreqIgnored=2,
epoches=20,
noOfWorkers=20,
learningRate=0.025,
distributedMem=1,
logFilename=None,
utilObj=None):
if (utilObj != None):
self.__util = utilObj
elif (logFilename != None):
self.__util = Utilities.Utility()
self.__util.setupLogFileLoc(logFilename)
self.__util.logDebug(self.__util.DOC2VECFRAME, 'Initialising doc2vec')
self.__util.startTimeTrack()
# self.__model = Doc2Vec(hashfxn=customHash.customHash, size=vectordim, min_count=wordFreqIgnored, iter=epoches, workers=noOfWorkers, alpha=learningRate, dm=distributedMem)
self.__model = Doc2Vec(size=vectordim,
min_count=wordFreqIgnored,
iter=epoches,
workers=noOfWorkers,
alpha=learningRate,
dm=distributedMem)
self.__util.logDebug(
self.__util.DOC2VECFRAME,
'Initialising doc2vec completed ' + self.__util.stopTimeTrack())
def __init__(self, utilObj=None, logFile=None):
if (utilObj != None):
self._util = utilObj
elif (logFile != None):
self._util = Utilities.Utility()
self._util.setupLogFileLoc(logFile)
pass
def __init__(self,
locationOfCVs=None,
utilObj=None,
logFilename=None,
**kwargs):
if (utilObj != None):
self.util = utilObj
elif (logFilename != None):
self.util = Utilities.Utility()
self.util.setupLogFileLoc(logFilename)
self.cvDataframe = pd.DataFrame(
columns=('filename', 'content', 'vector', 'proba',
'label')) # Structure of the csv to be saved
self.predictionModelType = kwargs['predictionModelType']
if (self.predictionModelType == 'topic'):
self.ldaModelFilename = kwargs['ldaModelFilename']
elif (self.predictionModelType == 'sim'):
self.vectorSpaceModelType = kwargs['vectorSpaceModelType']
self.vectorSpaceModelFilename = kwargs['vectorSpaceModelFilename']
self.predictionTrainingFilename = kwargs[
'predictionTrainingFilename']
else:
self.vectorSpaceModelType = kwargs['vectorSpaceModelType']
self.vectorSpaceModelFilename = kwargs['vectorSpaceModelFilename']
self.predictionTrainingFilename = kwargs[
'predictionTrainingFilename']
self.saveResultsFilename = kwargs['saveResultsFilename']
self.locationOfCVs = locationOfCVs
pass
def __init__(self):
super(Cognitive, self).__init__()
self.api = cognitive_face
file_path = Utilities.absolute_path(__file__, 'secret.json')
with open(file_path) as data_file:
data = json.load(data_file)
KEY = data['secret']
self.api.Key.set(KEY)
def register(self, data_path):
"""Register a group of people specified in data folder and train
Args:
data_path (string): train data folder
"""
if data_path == 'default':
data_path = Utilities.train_data_path()
if Utilities.file_exists(data_path):
group = ModelFactory.registered_group()
group.save()
for alias_name in os.listdir(data_path):
# Ignore gitkeep file and collect data from all folders
if alias_name != '.gitkeep':
logger.log('Registering %s...' % alias_name)
Person.register(group, alias_name)
# After everything is done, call api to train newly created group
self.train()
def __init__(self, logFilename=None, utilObj=None, vectordim=100, wordFreqIgnored=1,epoches=200, noOfWorkers=10,learningRate=0.025):
if (utilObj!=None):
self.util=utilObj
elif(logFilename!=None):
self.util = Utilities.Utility()
self.util.setupLogFileLoc(logFilename)
self.model = Doc2Vec(size=vectordim, min_count=wordFreqIgnored, iter=epoches, workers=noOfWorkers,alpha=learningRate)
self.util.logDebug('D2V', 'Initialising D2V')
self.util.startTimeTrack()
self.util.logDebug('D2V','Initialising D2V completed ' + self.util.stopTimeTrack())
def __init__(self, logFile=None, utilObj=None, **kwargs):
if (utilObj != None):
self._util = utilObj
elif (logFile != None):
self._util = Utilities.Utility()
self._util.setupLogFileLoc(logFile)
self._util.startTimeTrack()
self.keywordArgs=kwargs
print(self.keywordArgs)
def __init__(self, logFilename=None, utilObj=None, maxDim=None, ngram=1):
if (utilObj!=None):
self.util=utilObj
elif(logFilename!=None):
self.util = Utilities.Utility()
self.util.setupLogFileLoc(logFilename)
self.util.logDebug('TFIDF', 'Initialising TFIDF')
self.util.startTimeTrack()
self.model=TfidfVectorizer(max_features=maxDim, ngram_range=(1,ngram))
self.util.logDebug('TFIDF',
'Initialising TFIDF completed ' + self.util.stopTimeTrack())
pass
def __init__(self, camera):
"""Initialization
Args:
camera (int): index of camera to use
"""
self.cascade_file = Utilities.absolute_path(__file__, 'cascade.xml')
self.face_cascade = cv2.CascadeClassifier(self.cascade_file)
self.video_capture = cv2.VideoCapture(camera)
self.video_capture.set(cv2.cv.CV_CAP_PROP_FRAME_WIDTH, 640)
self.video_capture.set(cv2.cv.CV_CAP_PROP_FRAME_HEIGHT, 480)
self.faces_captured = 1
def register(group, alias):
data_path = Utilities.absolute_path(Utilities.train_data_path(), alias)
with open(os.path.join(data_path, 'name.txt'), 'r') as file:
name = file.read()
person = group.person_with_name(name)
person.alias = alias
create_voice_thread = threading.Thread(
name="create_voice",
target=Person.__create_voice,
args=(name, alias))
person.save()
save_faces_thread = threading.Thread(name="save_faces",
target=Person.__save_faces,
args=(person, data_path))
processes = ProcessParallel(create_voice_thread, save_faces_thread)
processes.fork_threads()
processes.start_all()
# Wait until all threads are done
processes.join_all()
def __init__(self, logFile=None, utilObj=None):
"""
:param ldaModel:
:param topicMapping:
:param labelledTestSamples: Can be any features since the vector doesn't matter.
"""
if (utilObj != None):
self._util = utilObj
elif (logFile != None):
self._util = Utilities.Utility()
self._util.setupLogFileLoc(logFile)
self._util.startTimeTrack()
def __init__(self, type=None, utilObj=None, utilName=None):
if (utilObj!=None):
self.util=utilObj
elif(utilName!=None):
selfutil = Utilities.Utility()
self.util.setupLogFileLoc(utilName)
if (type==self.TYPE_SVM):
self.classifier=SVC()
elif(type==self.TYPE_MULTINORM_NAIVEBAYES):
self.classifier=MultinomialNB()
pass
def __init__(self, logFile=None, utilObj=None, classifierType=None, classifierParams=None):
if (utilObj != None):
self._util = utilObj
elif (logFile != None):
self._util = Utilities.Utility()
self._util.setupLogFileLoc(logFile)
self._util.startTimeTrack()
if (classifierType==self.TYPE_SVM):
self._model = CalibratedClassifierCV(base_estimator=LinearSVC())
# self._model = LinearSVC()
self._type=classifierType
elif (classifierType==self.TYPE_LOG):
self._model = LogisticRegression(solver='newton-cg',n_jobs=30)
self._type = classifierType
elif (classifierType==self.TYPE_SVC):
self._model = SVC(C=5,kernel='linear')
self._type = classifierType
elif (classifierType==self.TYPE_MLP):
self._model = MLPClassifier()
self._type = classifierType
elif (classifierType==self.TYPE_NB):
self._model = MultinomialNB()
self._type = classifierType
elif (classifierType==self.TYPE_NB_GAUSSIAN):
self._model = GaussianNB()
self._type = classifierType
elif (classifierType == self.TYPE_TREE):
self._model = DecisionTreeClassifier()
self._type = classifierType
elif (classifierType == self.TYPE_XGBOOST):
self._model = XGBClassifier(max_depth=3, learning_rate=0.001, n_estimators=100, silent=False, objective='binary:logistic', booster='gbtree', n_jobs=32, nthread=None, gamma=0, min_child_weight=1, max_delta_step=0, subsample=1, colsample_bytree=1, colsample_bylevel=1, reg_alpha=0, reg_lambda=1, scale_pos_weight=1, base_score=0.5, random_state=0, seed=None, missing=None)
self._type = classifierType
elif (classifierType == self.TYPE_SEQ):
self._model = KerasWrapper(utilObj=self._util)
self._type = classifierType
elif (classifierType == self.TYPE_DUMMY):
try:
self._model = DummyClassifier()
self._type = classifierType
except Exception as error:
print(error)
else:
self._util.logError('MLClassifierPredictionModel','An instance of the classifier needs to be passed to this class...exiting')
exit(-1)
def create_voice(text, alias):
sounds_dir = Utilities.voices_path()
save_dir = Utilities.absolute_path(sounds_dir, alias + '.mp3')
if os.path.exists(save_dir) is False:
headers = {
'Content-Type': 'application/json',
'api_key': 'b5db987b5a944ff78097d435a5a564dc'
}
try:
response = requests.request(
'POST',
'http://api.openfpt.vn/text2speech/v3',
headers=headers,
data=text
)
if response.status_code not in (200, 202):
print response.status_code
if response.text:
result = response.json()
if result['async']:
wget.download(result['async'], save_dir)
except Exception as e:
print e
def __init__(self, logFile=None, utilObj=None, ldaModelFilename=None):
"""
:param ldaModel:
:param topicMapping:
:param labelledTestSamples: Can be any features since the vector doesn't matter.
"""
if (utilObj != None):
self._util = utilObj
elif (logFile != None):
self._util = Utilities.Utility()
self._util.setupLogFileLoc(logFile)
self._model = LDA.LDA(utilObj=self._util)
if (ldaModelFilename != None):
self._model.loadModel(ldaModelFilename)
_trained = True
pass
def start_capturing(self, is_register):
"""Start the camera, look for faces and write to file when faces are captured
Args:
is_register (bool): determine if the session is
registering or identifying
If this argument is true then camera will capture 3
photos of user. Otherwise, only 1 photo is captured
"""
if is_register is False:
self.faces_captured = DEFAULT_FACES - 1
while self.faces_captured < DEFAULT_FACES:
# Read frame from video capture
return_code, frame = self.video_capture.read()
# Use Haar cascade to detect faces in captured frame
faces = self.face_cascade.detectMultiScale(
cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY),
scaleFactor=1.1,
minNeighbors=5,
minSize=(30, 30),
flags=cv2.cv.CV_HAAR_SCALE_IMAGE
)
# Looping through captured faces and write to file
for (x, y, w, h) in faces:
# All captured images will be saved in tmp folder in
# core/face_recognizer
filePath = Utilities.absolute_path(
__file__, 'tmp/face%d.jpg' % self.faces_captured)
self.faces_captured += 1
cv2.imwrite(filePath, frame)
if is_register:
# Sleep 2 seconds so user can change face orientation
time.sleep(2)
break
if cv2.waitKey(1) & 0xFF == ord('q'):
self.stop_capturing()
break
# cv2.imshow('Video', frame)
self.stop_capturing()
appFolderLocation = None
if (len(sys.argv) == 11):
listOfAppD2vFoldersOrderByLabelStr = sys.argv[8]
listOfAppD2vFoldersOrderByLabel = listOfAppD2vFoldersOrderByLabelStr.split(
',')
listOfLabelsOrderByLabelStr = sys.argv[9]
listOfLabelsOrderByLabel = list(
map(int, listOfLabelsOrderByLabelStr.split(',')))
appFolderLocation = sys.argv[10]
print('listOfAppD2vFoldersOrderByLabelStr:',
listOfAppD2vFoldersOrderByLabelStr)
print('listOfLabelsOrderByLabelStr:', listOfLabelsOrderByLabelStr)
print('appFolderLocation:', appFolderLocation)
util = Utilities.Utility()
util.setupLogFileLoc(logFile=logFile)
util.setupTokenizationRules(tokenRules)
predictionModel = MLClassifierPredictionModel.MLClassifierPredictionModel(
utilObj=util, classifierType=classifierType)
predictionModel.train(sampleLabelledTrainingFilename)
predictionModel.loadXYtest(sampleLabelledTestFilename)
eval = Evaluator.Evaluator(utilObj=util)
predictionModel.evaluate(approach_vsm, eval)
if (len(sys.argv) == 11):
evalHeu = HeuristicEvaluator.HeuristicEvaluator(
utilObj=util,
listOfAppD2vFoldersOrderByLabel=listOfAppD2vFoldersOrderByLabel,
listOfLabelsOrderByLabel=listOfLabelsOrderByLabel,
class Evaluator():
SCORE_ACCURACY='acc'
SCORE_CONFUSIONMATRIX='confmatrix'
SCORE_CLASSREPORT='class'
SCORE_F1='f1'
SCORE_PRECISION='precision'
SCORE_RECALL='recall'
SCORE_F1_PERCLASS='f1_perclass'
SCORE_PRECISION_PERCLASS='precision_perclass'
SCORE_RECALL_PERCLASS='recall_perclass'
__logFile=''
util=Utilities.Utility()
def __init__(self, logFile=None, utilObj=None):
if (utilObj != None):
self.util = utilObj
elif (logFile != None):
self.util = Utilities.Utility()
self.util.setupLogFileLoc(logFile)
def generateGraphicalConfusionMatrix(self, array):
import seaborn as sn
import matplotlib.pyplot as plt
df_cm = pd.DataFrame(array, index=[i for i in "ESPW"],
columns=[i for i in "ESPW"])
plt.figure(figsize=(10, 7))
confMatrix=sn.heatmap(df_cm, annot=True, cmap="YlGnBu" )
return plt
pass
def generateSummary(self, folderpath):
from decimal import Decimal
metricDict=defaultdict(list)
for metricFile in sorted(glob.iglob(folderpath + '/*.metric')):
tokens=metricFile.split('/')[-1].split('_')
approach=tokens[0]
vsm=tokens[1]
metric=tokens[2].split('.')[0]
value=str(round(Decimal(self.util.readFileContent(metricFile)),2))
recordTuple=(approach,vsm,value)
if(metric in metricDict):
tupleList=metricDict[metric]
tupleList.append(recordTuple)
metricDict[metric]=tupleList
else:
tupleList=[recordTuple]
metricDict[metric] = tupleList
metricTuple=[]
for key,value in metricDict.items():
metricResultsDF = pd.DataFrame(columns=(['VSM']))
metricResultsDF = metricResultsDF.set_index("VSM")
for tupleList in value:
approach=tupleList[0]
vsm = tupleList[1]
metricValue = tupleList[2]
metricResultsDF.set_value(vsm, approach, metricValue)
print(metricResultsDF)
metricResultsDF.to_csv(folderpath+'/'+key+'.summary',header=True,index_label='VSM',index=True,mode='w')
def printSummary(self):
pass
def score(self,y=None, ypred=None,type=SCORE_ACCURACY, filename=None, **kwargs):
results=None
if(type==self.SCORE_ACCURACY):
# print(y)
# print(ypred)
results=accuracy_score(y,ypred)
self.util.saveStringToFile(results,filename=filename+'_ACC.metric')
elif(type==self.SCORE_CONFUSIONMATRIX):
print('y:',y)
print('ypred:', ypred)
results=confusion_matrix(y,ypred)
print('results:',results)
try:
import seaborn as sn
import matplotlib.pyplot as plt
snMatrix=self.generateGraphicalConfusionMatrix(results)
snMatrix.savefig(filename + '_CONF.png')
except Exception as error:
print(error)
self.util.logError('Evaluator','Graphical version of confusion matrix cannot be generated with xScreen and tkinter support on python')
results=np.array2string(results)
self.util.saveStringToFile((results), filename=filename + '_CONF.CONFU')
elif(type==self.SCORE_CLASSREPORT):
results=classification_report(y,ypred)
# self.util.saveStringToFile(results, filename=filename + '_CLASS.metric')
elif(type==self.SCORE_F1_PERCLASS):
results=f1_score(y,ypred, average=None)
counter=0
for result in results:
self.util.saveStringToFile(results[counter], filename=filename + '_F1-'+str(counter)+'.metric')
counter=counter+1
elif(type==self.SCORE_PRECISION_PERCLASS):
results=precision_score(y,ypred, average=None)
counter=0
for result in results:
self.util.saveStringToFile(results[counter], filename=filename + '_PREC-'+str(counter)+'.metric')
counter=counter+1
elif(type==self.SCORE_RECALL_PERCLASS):
results=recall_score(y,ypred,average=None)
counter=0
for result in results:
self.util.saveStringToFile(results[counter], filename=filename + '_RECALL-'+str(counter)+'.metric')
counter=counter+1
elif(type==self.SCORE_F1):
results=f1_score(y,ypred, average='macro')
self.util.saveStringToFile(results, filename=filename + '_F1.metric')
elif(type==self.SCORE_PRECISION):
results=precision_score(y,ypred, average='macro')
self.util.saveStringToFile(results, filename=filename + '_PREC.metric')
elif(type==self.SCORE_RECALL):
results=recall_score(y,ypred,average='macro')
self.util.saveStringToFile(results, filename=filename + '_RECALL.metric')
return results
def _getFirstString(self,myDelimitedStr=None, delimiter=','):
"""
Return the first string from a set of delimited strings
:param categoryStr:
:param delimiter:
:return:
"""
try:
results=myDelimitedStr.split(delimiter)[0]
except Exception as error:
results=''
return (results)
def evaluateHeuristic(self, resultsDatasetFilename=None, heuristicsFilename=None, appd2vEduFolder=None, appd2vSkillsFolder=None, appd2vPersonalDetailsFolder=None):
"""
resultsDataset contains the categorised results from Test phase 2.
This will have information on appid, clientid, cvd2vfilename,and its inferred categories (edu, skills, personaldetails) and corresponding scores.
appd2vEduFolder,appd2vSkillsFolder and appd2vPersonalDetailsFolder will contain the appd2v files which are named app_xx_yyyy.d2v.
For each row in resultsDataset,
- pull the content of cvd2vfilename.
- pull the content of appd2v[category]Folder/app_xx_yyyy.d2v.
- Strip the stop words
compare every word in both content,
- if the number of words identical hit a certain threshold, then HIT.
- else, MISS
Add this HIT/MISS into the resultsDataset csv as a new column.
Add identical words as a new column
:param resultsDatasetFilename: Names of files seperated by ';'
:param heuristicsFilename: Name of the file to save results in
:param appd2vEduFolder: Folder that contain the appd2v files which are named app_xx_yyyy.d2v.
:param appd2vSkillsFolder: Folder that contain the appd2v files which are named app_xx_yyyy.d2v.
:param appd2vPersonalDetailsFolder: Folder that contain the appd2v files which are named app_xx_yyyy.d2v.
:return:
"""
filecounter=0
resultsDatasetFilenames=resultsDatasetFilename.split(';')
fullResults = pd.DataFrame(columns=(
'appid', 'clientid', 'cvd2vfilename', 'categories', 'scores', 'heuristics', 'heuristics_reason','content'))
hitCounter=0
MissCounter = 0
for resultsDatasetFilename in resultsDatasetFilenames:
#To confirm if the filename contains headers.
self.util.logDebug('Evaluator-evaluateFromDataset', 'Reading ' + resultsDatasetFilename)
resultsDF=None
resultsDF=pd.read_csv(resultsDatasetFilename, header=None)
#Should load as 'appid', 'clientid', 'cvd2vfilename','content', 'categories', 'scores'
self.util.logDebug('Evaluator-evaluateFromDataset', 'Processing...')
counter=0
errcounter=0
for index, row in resultsDF.iterrows():
# try:
clientid=row[1]
clientid=(str(clientid)).zfill(3)
appid=row[0]
cvd2vFullpath=row[2]
categories=row[4]
scores=row[5]
content=row[3]
#The heuristics will only take the category with highest score.
category=self._getFirstString(myDelimitedStr=categories, delimiter=':')
score=self._getFirstString(myDelimitedStr=scores, delimiter=':')
cvd2vContent = open(cvd2vFullpath, 'r').read()
cvd2vContentTokens=self.util.tokenize(cvd2vContent) #This is to be used for comparison.
heuristics_reason=''
appd2vFilename=''
#Based on the matched category, pull the relevant app_xx_yyyy.d2v file.
if (category==self.util.LOOKUP_CAT_EDU):
appd2vFilename=appd2vEduFolder+'/'+'app_'+str(clientid)+'_'+str(appid)+'.'+ self.util.LOOKUP_EXT_APPD2V
elif(category==self.util.LOOKUP_CAT_SKILLS):
appd2vFilename = appd2vSkillsFolder + '/' + 'app_' + str(clientid) + '_' + str(
appid) + '.' + self.util.LOOKUP_EXT_APPD2V
elif (category == self.util.LOOKUP_CAT_PERSONALDETAILS):
appd2vFilename = appd2vPersonalDetailsFolder + '/' + 'app_' + str(clientid) + '_' + str(
appid) + '.' + self.util.LOOKUP_EXT_APPD2V
if (os.path.exists(appd2vFilename)==True):
appd2vContent = open(appd2vFilename, 'r').read()
else:
appd2vContent=''
heuristics_reason='FILE_NOT_FOUND_IN_CATEGORY: '+appd2vFilename
appd2vContentTokens = self.util.tokenize(appd2vContent) # This is to be used for comparison.
identicals=set(appd2vContentTokens) & set(cvd2vContentTokens)
# if(len(identicals)>0):
# heuristics_reason=self.util.tokensToStr(identicals)
heuristics='MISS'
if(category==self.util.LOOKUP_CAT_EDU and len(identicals)>=self.util.THRES_EDU):
heuristics='HIT'
heuristics_reason = self.util.tokensToStr(identicals)
elif(category==self.util.LOOKUP_CAT_SKILLS and len(identicals)>=self.util.THRES_SKILLS):
heuristics = 'HIT'
heuristics_reason = self.util.tokensToStr(identicals)
elif (category == self.util.LOOKUP_CAT_PERSONALDETAILS and len(identicals) >= self.util.THRES_PERSONALDETAILS):
## For personal details, can be more restrictive by limiting to words not in English Dictionary
identicals=self.util.returnNonEnglishDictWords(identicals)
if(len(identicals)>self.util.THRES_PERSONALDETAILS):
heuristics = 'HIT'
heuristics_reason = self.util.tokensToStr(identicals)
# print(heuristics)
currentRow = pd.DataFrame(data={'appid': [appid], 'clientid': [clientid],'cvd2vfilename': [resultsDatasetFilename], 'categories': [categories],'scores': [scores], 'heuristics': [heuristics],'heuristics_reason': [heuristics_reason],'content': [cvd2vContent]})
fullResults=fullResults.append(currentRow)
counter = counter + 1
if counter%100 ==0:
self.util.logDebug('Evaluator-evaluateFromDataset',
str(counter) + ' files completed with ' + str(errcounter) + ' errors.')
self.util.logDebug('Evaluator-evaluateFromDataset', 'Saving!')
fullResults.to_csv(heuristicsFilename.split('.')[0]+'_'+str((filecounter)).zfill(2)+'.csv', ',', mode='a', header=False, index=False,
columns=['appid', 'clientid', 'cvd2vfilename', 'categories', 'scores', 'heuristics', 'heuristics_reason', 'content'])
fullResults = fullResults[0:0]
# except Exception as error:
# errcounter=errcounter+1
# self.util.logDebug('Evaluator-evaluateFromDataset', 'Error encountered: ' + repr(error))
self.util.logDebug('Evaluator-evaluateFromDataset', 'Final save!')
fullResults.to_csv(heuristicsFilename.split('.')[0]+'_'+str((filecounter)).zfill(2)+'.csv', ',', mode='a', header=False, index=False,
columns=['appid', 'clientid', 'cvd2vfilename', 'categories', 'scores', 'heuristics',
'heuristics_reason', 'content'])
filecounter=filecounter+1
# # TestA phase 1xa
# # python3 Evaluator.py '/u01/bigdata/03a_01b_test/cvd2v/test2/heuEval.log' '/u01/bigdata/03a_01b_test/cvd2v/test2/TestA_32_userlabelled_results_1000.csv;/u01/bigdata/03a_01b_test/cvd2v/test2/TestA_32_userlabelled_results_2000.csv;/u01/bigdata/03a_01b_test/cvd2v/test2/TestA_32_userlabelled_results_2759.csv' '/u01/bigdata/03a_01b_test/cvd2v/test2/heuEval.csv' '/u01/bigdata/01b_d2v/032/edu/doc2vecEdu' '/u01/bigdata/01b_d2v/032/skills/doc2vecSkills' '/u01/bigdata/01b_d2v/032/personaldetails/doc2vecPersonalDetails'
# if __name__ == "__main__":
# if(len(sys.argv)==7):
# logFile = sys.argv[1]
# resultsDatasetFilename = sys.argv[2]
# heuristicsFilename=(sys.argv[3])
# appd2vEduFolder=(sys.argv[4])
# appd2vSkillsFolder=sys.argv[5]
# appd2vPersonalDetailsFolder=sys.argv[6]
# print('Logging to ', logFile)
# print('resultsDatasetFilename',resultsDatasetFilename)
# print('heuristicsFilename', heuristicsFilename)
# print('appd2vEduFolder', appd2vEduFolder)
# print('appd2vSkillsFolder', appd2vSkillsFolder)
# print('appd2vPersonalDetailsFolder', appd2vPersonalDetailsFolder)
#
# eval=Evaluator(logFile)
# eval.evaluateHeuristic(resultsDatasetFilename=resultsDatasetFilename, heuristicsFilename=heuristicsFilename, appd2vEduFolder=appd2vEduFolder, appd2vSkillsFolder=appd2vSkillsFolder, appd2vPersonalDetailsFolder=appd2vPersonalDetailsFolder)
# else:
# print('Arguments incorrect')
# e=Evaluator()
# e.generateSummary('/u01/bigdata/02d_d2vModel1/features')
def __init__(self,logfile=None):
self._logfile=logfile
self._util = Utilities.Utility()
self._util.setupLogFileLoc(logfile)