def naiveBayesMC():
binSize = 10;
maxPrice = 100;
dataSplit = 0.70;
print "Loading items... "
allItems = getItems(db,filteritems=True,complete=True,sold=True,genre=ROCK);
[trainItems,testItems] = splitItemSet(allItems,dataSplit);
bins = generateBinArray(binSize,maxPrice);
[actualFinalPrice,actualFinalPriceBinned] = getFinalPrices(testItems,bins);
##Only have to make these onece
orderedWordList = generateOrderedWordList(allItems,lengthcutoff=5,frequencycutoff=1);
[testMatrix,testCategory] = generateMatrixData(orderedWordList,generateItemTitleList(testItems,orderedWordList));
predictedFinalPrices = [-1]*len(testItems);
for priceCutOff in bins:
print "Price cut off: ", priceCutOff
[phi_k_unsold,phi_k_sold,p_y0,p_y1] = trainOnData(trainItems,orderedWordList,priceCutOff);
[testingSetPredictions,prob_sell,prob_wontSell] = makePredictions(testMatrix,phi_k_sold,phi_k_unsold,p_y0,p_y1); #predicted category for testItems [0,1]
predictedFinalPrices = updatePredictedFinalPrice(testItems,predictedFinalPrices,testingSetPredictions,priceCutOff,binSize,bins);
for i in range(len(predictedFinalPrices)):
if predictedFinalPrices[i] == -1: predictedFinalPrices[i] = bins[-1];
for i in range(len(predictedFinalPrices)):
print i,testItems[i][0],testItems[i][START_PRICE],testItems[i][END_PRICE],getBinOf(bins,float(testItems[i][START_PRICE])), actualFinalPriceBinned[i],predictedFinalPrices[i],"\t\t",testItems[i][TITLE]
print "Classification error on testing set is: ", classificationError(predictedFinalPrices,actualFinalPriceBinned);
def runClassificationErrorTest():
fail = False;
p2 = [0,1,0];
a2 = [0,1,1];
if (m.classificationError([0],[0]) != 0): fail = True;
if (m.classificationError([0],[1]) != 1): fail = True;
if (m.classificationError([0],[-1]) != 1): fail = True;
if (m.classificationError([0,0],[0,0]) != 0): fail = True;
if (m.classificationError([0,0],[0,1]) != 0.5): fail = True;
if (m.classificationError([0,0],[1,1]) != 1): fail = True;
if (fail):
print "Classification Error Function has failed!";
print "Items in testing set: ",len(testItems);
orderedWordList = generateOrderedWordList(allItems,frequencycutoff = 4);
print "Training on data set ..."
[phi_k_unsold,phi_k_sold,p_y0,p_y1] = trainOnData(trainItems,orderedWordList,0)
printTopWords(phi_k_sold,phi_k_unsold,orderedWordList)
print "Preparing test data ... "
itemTitleList = generateItemTitleList(testItems,orderedWordList);
[testMatrix,testCategory] = generateMatrixData(orderedWordList,itemTitleList);
print "Making predictions ... "
[testingSetPredictions,prob_sell,prob_wontSell] = makePredictions(testMatrix,phi_k_sold,phi_k_unsold,p_y0,p_y1,uniformPrior=False);
print "Classification error on testing set is: ", classificationError(testingSetPredictions,testCategory);
def printTopWords(phi_k_sold,phi_k_unsold,orderedWordList):
rating = [];
for i in range(len(phi_k_sold)):
rating.append(np.log(phi_k_sold[i] / phi_k_unsold[i]) );
ratingsSorted = [orderedWordList[i[0]] for i in sorted(enumerate(rating), key=lambda x:x[1])]
reverseratingsSorted = [orderedWordList[i[0]] for i in sorted(enumerate(rating), key=lambda x:x[1])]
reverseratingsSorted.reverse();
print
print "Highest rated words: ", ratingsSorted[0:30]
print "Lowest rated: ",reverseratingsSorted[0:30]
##
