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

removeUselessSeries and stronger linear seperation

Browse Source
This commit is contained in:
Dominik Moritz Roth 2021-09-26 12:46:30 +02:00
parent 2bb2e15b73
commit f1c887275c
1 changed files with 17 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
caliGraph.py
View File

@ -344,6 +344,12 @@ def removeUselessTags(G, minUnread=1):
if foundUnread < minUnread: if foundUnread < minUnread:
G.remove_node(n) G.remove_node(n)
def removeUselessSeries(G, minSco=0):
for n in list(G.nodes):
node = G.nodes[n]
if node['t'] == 'series':
if len(G.adj[n]) < 2 or node['score'] < minSco:
G.remove_node(n)
def scoreOpinions(G, globMu, globStd): def scoreOpinions(G, globMu, globStd):
for n in list(G.nodes): for n in list(G.nodes):
@ -386,7 +392,8 @@ def scoreUnread(G, globMu, globStd):
ws.append(getWeightForType('sigma')) ws.append(getWeightForType('sigma'))
feedbacks.append(node['se']) feedbacks.append(node['se'])
ws.append(getWeightForType('se')) ws.append(getWeightForType('se'))
#node['score'] = sum([fb*w for fb, w in zip(feedbacks, weights)])/len(feedbacks) feedbacks.append(globMu)
ws.append(getWeightForType('bias'))
node['score'] = sum([fb*w for fb, w in zip(feedbacks, ws)])/sum(ws) node['score'] = sum([fb*w for fb, w in zip(feedbacks, ws)])/sum(ws)
else: else:
node['score'] = globMu + errorFac*globStd + len(feedbacks)*0.0000000001 node['score'] = globMu + errorFac*globStd + len(feedbacks)*0.0000000001
@ -644,13 +651,17 @@ def recommendNBooks(G, mu, std, n, removeTopListsB=True, removeUselessRecommende
pruneTags(G, 12) pruneTags(G, 12)
pruneAuthorCons(G, int(n/5)) pruneAuthorCons(G, int(n/5))
pruneRecommenders(G, 12 - min(5, n/20)) pruneRecommenders(G, 12 - min(5, n/20))
removeUselessSeries(G, mu)
removeUselessTags(G) removeUselessTags(G)
if removeTopListsB: if removeTopListsB:
removeTopLists(G) removeTopLists(G)
removeDangling(G, alsoBooks=True) removeDangling(G, alsoBooks=True)
removeKeepBest(G, n+math.ceil(n/20)+3, maxDistForRead=1.5) removeKeepBest(G, n+math.ceil(n/20)+3, maxDistForRead=1.5)
removeUselessReadBooks(G)
removeEdge(G) removeEdge(G)
removeKeepBest(G, n+1, maxDistForRead=1.25)
removeUselessSeries(G, mu)
removeUselessTags(G)
removeUselessReadBooks(G)
removeKeepBest(G, n, maxDistForRead=1.25) removeKeepBest(G, n, maxDistForRead=1.25)
scaleBooksByRating(G) scaleBooksByRating(G)
@ -775,6 +786,7 @@ def evaluateFitness(books):
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
linSepLoss = []
errSq = [] errSq = []
mu, sigma = genScores(G, books) mu, sigma = genScores(G, books)
for b in G.nodes: for b in G.nodes:
@ -794,9 +806,10 @@ def evaluateFitness(books):
elif score < 0.0: elif score < 0.0:
boundsLoss += (score)**2 boundsLoss += (score)**2
# reward seperation linearly # reward seperation linearly
boundsLoss -= abs(score - mu)/10 linSepLoss.append(abs(score - mu))
regressionLoss = sum([(1-w)**2 for w in weights.values()]) regressionLoss = sum([(1-w)**2 for w in weights.values()])
return sum(errSq)/len(errSq) + regressionLoss/100 + boundsLoss/100 print(sum(errSq)/len(errSq), 0.005*regressionLoss, 0.2*boundsLoss/len(ratedBooks), 0.5*sum(linSepLoss)/len(linSepLoss))
return sum(errSq)/len(errSq) + 0.005*regressionLoss + 0.2*boundsLoss/len(ratedBooks) - 0.5*sum(linSepLoss)/len(linSepLoss)
def train(gamma = 1): def train(gamma = 1):
global weights global weights