def getRowsNumber(self):
# predefined where are numerical and subject columns
dataset = Dataset(self.datasetPath, 1, [2])
if self.checkDataset(dataset.df) == False:
return -1
return dataset.df.size
def getScores(self):
# predefined where are numerical and subject columns
dataset = Dataset(self.datasetPath, 1, [2])
if self.checkDataset(dataset.df) == False:
return -1
sem = Semantification(dataset)
scores = {}
scores['columnResults'] = sem.columnsResultsKS
scores['rowResults'] = sem.rowPredictions
scores['finalResults'] = sem.finalResults
return scores
def getScores(self):
# predefined where are numerical and subject columns
dataset = Dataset(self.datasetPath, 0, [2])
if self.checkDataset(dataset.df) == False:
return -1
correctProperty = self.getCorrectPrediction()
# overwriting column type with a correct type & cell types
dataset.subjectColumn.columnTypes = [(correctProperty,1)]
for cell in dataset.subjectColumn.cellPredictions:
#print cell.cell
cell.uri = cell.cell
cell.types = correctProperty
sem = Semantification(dataset)
scores = {}
scores['columnResults'] = sem.columnsResultsKS
scores['rowResults'] = sem.rowPredictions
scores['finalResults'] = sem.finalResults
#self.checkCorrect(scores['finalResults'])
return scores
def getColumnMapping(self):
dataset = Dataset(self.datasetPath , 0, [2])
if self.checkDataset(dataset.df) == False:
return -1
return list(dataset.df.columns.values)
from approach.Dataset import Dataset
#dataset = Dataset('10786782_0_7941448888047609465.csv')
dataset = Dataset('96960685_0_6886906070865701391.csv')
print "subject columns is " + str(dataset.subjectColumnId)
print "numerical column are [2] "
print dataset.numericalColumnsIds
print dataset.columnsWithMappingsIds
dataset = Dataset('10786782_0_7941448888047609465.csv')
print "subject columns is " + str(dataset.subjectColumnId)
print "numerical column is "
print dataset.numericalColumnsIds
print dataset.columnsWithMappingsIds
dataset = Dataset('52340077_0_7623033473986759010.csv')
print "subject columns is " + str(dataset.subjectColumnId)
print "numerical column is "
print dataset.numericalColumnsIds
print dataset.columnsWithMappingsIds
from approach.Dataset import Dataset from approach.config.paths import * from approach.config.imports import * d = Dataset(path) #print d.name #print "Rows: " + str(d.noRows) #print "Column: " + str(d.noColumns) #print "Headers are: " #print d.printHeaders() #for column in d.columns: #print column.name # print column.type
from approach.Dataset import Dataset
from approach.config.paths import *
from approach.config.imports import *
for fn in os.listdir(datasetsPath):
print "______________________________________________"
print fn
dataset = Dataset(fn)
from approach.Dataset import Dataset
#dataset = Dataset('10786782_0_7941448888047609465.csv')
#dataset = Dataset('96960685_0_6886906070865701391.csv')
dataset = Dataset('56834172_0_5710924050177414995.csv')
from approach.Dataset import Dataset
from approach.Semantification import Semantification
from approach.config.paths import *
from approach.config.imports import *
#for fn in os.listdir(datasetsPath):
# sem = Semantification(fn)
# sem.getRowPredictions()
#sem = Semantification('56834172_0_5710924050177414995.csv')
dataset = Dataset('/Users/emiliakacprzak/Code/papers/semantification/data/dbpediaTest.csv', 0, [1])
sem = Semantification(dataset)
print sem.columnsResultsKS
print "___"
print sem.rowPredictions
print "___"
print sem.finalResults