dodox/CaliGraph
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Better Recommendation Generation

Browse Source
This commit is contained in:
Dominik Moritz Roth 2022-02-01 12:22:13 +01:00
parent 6e64e76310
commit bd17ac4906
1 changed files with 11 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

23
caliGraph.py
View File

@ -634,12 +634,12 @@ def buildFullGraph(darkMode=False):
graphAddTopLists(G, books, darkMode=darkMode) graphAddTopLists(G, books, darkMode=darkMode)
graphAddSeries(G, books, darkMode=darkMode) graphAddSeries(G, books, darkMode=darkMode)
graphAddTags(G, books, darkMode=darkMode) graphAddTags(G, books, darkMode=darkMode)
runPagerank(G)
return G, books return G, books
def genScores(G, books): def genScores(G, books):
globMu, globStd = calcRecDist(G, books) globMu, globStd = calcRecDist(G, books)
runPagerank(G)
scoreOpinions(G, globMu, globStd) scoreOpinions(G, globMu, globStd)
scoreUnread(G, globMu, globStd) scoreUnread(G, globMu, globStd)
return globMu, globStd return globMu, globStd
@ -708,16 +708,16 @@ def recommendNBooks(G, mu, std, n, removeTopListsB=True, removeUselessRecommende
removeBad(G, mu+std/2, groups=['recommender']) removeBad(G, mu+std/2, groups=['recommender'])
removeKeepBest(G, int(n*2) + 5, maxDistForRead=2) removeKeepBest(G, int(n*2) + 5, maxDistForRead=2)
removeEdge(G) removeEdge(G)
removeHighSpanTags(G, 10) removeHighSpanTags(G, 8)
removeHighSpanReadBooks(G, 10) removeHighSpanReadBooks(G, 14)
removeDangling(G, alsoBooks=False) removeDangling(G, alsoBooks=False)
pruneRecommenders(G, 13) pruneRecommenders(G, 16)
pruneTags(G, 13) pruneTags(G, 9)
removeBad(G, mu, groups=['book']) removeBad(G, mu, groups=['book'])
removeUselessReadBooks(G) removeUselessReadBooks(G)
pruneTags(G, 12) pruneTags(G, 8)
pruneAuthorCons(G, int(n/5)) pruneAuthorCons(G, int(n/5)+3)
pruneRecommenders(G, 12 - min(5, n/20)) pruneRecommenders(G, 16 - min(4, n/20))
removeUselessSeries(G, mu) removeUselessSeries(G, mu)
removeUselessTags(G) removeUselessTags(G)
if removeTopListsB: if removeTopListsB:
@ -911,7 +911,6 @@ def evaluateFitness(books, debugPrint=False):
graphAddTopLists(G, books) graphAddTopLists(G, books)
graphAddSeries(G, books) graphAddSeries(G, books)
graphAddTags(G, books) graphAddTags(G, books)
runPagerank(G)
ratedBooks = [n for n in list(G.nodes) if 'rating' in G.nodes[n] and G.nodes[n]['rating'] != None] ratedBooks = [n for n in list(G.nodes) if 'rating' in G.nodes[n] and G.nodes[n]['rating'] != None]
boundsLoss = 0 boundsLoss = 0
@ -946,8 +945,8 @@ def evaluateFitness(books, debugPrint=False):
for g in gradient: for g in gradient:
gradient[g] /= len(errSq) gradient[g] /= len(errSq)
if debugPrint: if debugPrint:
print(sum(errSq)/len(errSq), 0.003*regressionLoss, 0.2*boundsLoss/len(ratedBooks), 1.0*sum(linSepLoss)/len(linSepLoss)) print(sum(errSq)/len(errSq), 0.001*regressionLoss, 0.1*boundsLoss/len(ratedBooks), 0.5*sum(linSepLoss)/len(linSepLoss))
fit = sum(errSq)/len(errSq) + 0.003*regressionLoss + 0.2*boundsLoss/len(ratedBooks) - 1.0*sum(linSepLoss)/len(linSepLoss) fit = sum(errSq)/len(errSq) + 0.001*regressionLoss + 0.1*boundsLoss/len(ratedBooks) - 0.5*sum(linSepLoss)/len(linSepLoss)
return fit, gradient return fit, gradient
def train(initGamma, full=True): def train(initGamma, full=True):
@ -992,7 +991,7 @@ def train(initGamma, full=True):
gamma = initGamma gamma = initGamma
if random.random() < 0.50: if random.random() < 0.50:
for wt in weights: for wt in weights:
weights[wt] = random.random() weights[wt] = random.random()*2-0.5
else: else:
weights = copy.copy(bestWeights) weights = copy.copy(bestWeights)
for wt in weights: for wt in weights: