def averagePrepare(product): #Create path of all the reviews allpath = "../../datasets/reviews_"+product+".csv" #Create paths of training and testing reviews trainpath = "train_"+product+".csv" testpath = "test_"+product+".csv" #Read reviews df_train = pandas.read_csv(trainpath) df_test = pandas.read_csv(testpath) df = pandas.read_csv(allpath) #Prepare the wordList for the product reviews = process.processReviews(df) wordList = process.preprocessWordlist(reviews) #Create the vectors of the bag of words trainX = np.array(process.formatAverageX(df_train, wordList)) testX = np.array(process.formatAverageX(df_test, wordList)) #Create the vectors of the evaluations trainY = np.array(process.formatY(df_train)) testY = np.array(process.formatY(df_test)) return [trainX, trainY, testX, testY]
def basePrepare(i):
#Create path of all the reviews
allpath = "../../datasets/reviews_all.csv"
#Create paths of training and testing reviews
testpath = "kFold/test_all_" + str(i) + ".csv"
#Read reviews
df = pandas.read_csv(allpath)
df_test = pandas.read_csv(testpath)
df_train = df.drop(df_test.index)
#Prepare the wordList for the product
reviews = process.processReviews(df)
wordList = process.preprocessWordlist(reviews)
#Create the vectors of the bag of words
trainX = np.array(process.formatBaseX(df_train, wordList))
testX = np.array(process.formatBaseX(df_test, wordList))
#Create the vectors of the evaluations
trainY = np.array(process.formatY(df_train))
testY = np.array(process.formatY(df_test))
return [trainX, trainY, testX, testY]
