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

Added a regression-loss (push weights towards 1)

Browse Source
This commit is contained in:
Dominik Moritz Roth 2021-09-24 19:12:09 +02:00
parent 9d6f37af45
commit 5a4e48d86c
1 changed files with 12 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
caliGraph.py
View File

@ -341,7 +341,7 @@ def scoreOpinions(G, globMu, globStd):
def scoreUnread(G, globMu, globStd): def scoreUnread(G, globMu, globStd):
for n in list(G.nodes): for n in list(G.nodes):
feedbacks = [globMu] feedbacks = [globMu]
weights = [getWeightForType('mu')] ws = [getWeightForType('mu')]
node = G.nodes[n] node = G.nodes[n]
if node['t'] == 'book': if node['t'] == 'book':
if node['rating'] == None: if node['rating'] == None:
@ -352,16 +352,16 @@ def scoreUnread(G, globMu, globStd):
w = getWeightForType(adjNode['t'], G[n][adj]['weight'] if 'weight' in G[n][adj] else 1) w = getWeightForType(adjNode['t'], G[n][adj]['weight'] if 'weight' in G[n][adj] else 1)
for fb in adjNode['feedbacks']: for fb in adjNode['feedbacks']:
feedbacks.append(fb) feedbacks.append(fb)
weights.append(w) ws.append(w)
if len(feedbacks): if len(feedbacks):
node['mean'], node['std'] = norm.fit(feedbacks) node['mean'], node['std'] = norm.fit(feedbacks)
node['se'] = globStd / math.sqrt(len(feedbacks)) node['se'] = globStd / math.sqrt(len(feedbacks))
feedbacks.append(node['std']) feedbacks.append(node['std'])
weights.append(getWeightForType('sigma')) ws.append(getWeightForType('sigma'))
feedbacks.append(node['se']) feedbacks.append(node['se'])
weights.append(getWeightForType('se')) ws.append(getWeightForType('se'))
node['score'] = sum([fb*w for fb, w in zip(feedbacks, weights)])/len(feedbacks) #node['score'] = sum([fb*w for fb, w in zip(feedbacks, weights)])/len(feedbacks)
#node['score'] = sum([fb*w for fb, w in zip(feedbacks, weights)])/sum(weights) node['score'] = sum([fb*w for fb, w in zip(feedbacks, ws)])/len(feedbacks)
else: else:
node['score'] = globMu + errorFac*globStd + len(feedbacks)*0.0000000001 node['score'] = globMu + errorFac*globStd + len(feedbacks)*0.0000000001
if 'series' in node: if 'series' in node:
@ -683,6 +683,7 @@ def waveFlow(G, node, n, dist, menge, firstEdge=False):
waveFlow(G, node, m, dist, menge, firstEdge=firstEdge) waveFlow(G, node, m, dist, menge, firstEdge=firstEdge)
def evaluateFitness(books): def evaluateFitness(books):
global weights
G = buildBookGraph(books) G = buildBookGraph(books)
graphAddAuthors(G, books) graphAddAuthors(G, books)
graphAddRecommenders(G, books) graphAddRecommenders(G, books)
@ -703,7 +704,8 @@ def evaluateFitness(books):
if rating > 10.0: if rating > 10.0:
errSq[-1] *= 1.5 errSq[-1] *= 1.5
G.nodes[m]['rating'] = rating G.nodes[m]['rating'] = rating
return sum(errSq) / len(errSq) regressionLoss = sum([(1-w)**2 for w in weights.values()])
return sum(errSq)/len(errSq) + regressionLoss/1000
def train(gamma = 0.1): def train(gamma = 0.1):
global weights global weights
@ -720,15 +722,15 @@ def train(gamma = 0.1):
if gamma < 0.01: if gamma < 0.01:
while random.random() < 0.5: while random.random() < 0.5:
attr = random.choice(w) attr = random.choice(w)
weights[attr] = -0.1+random.random()*1.5 weights[attr] = -0.1+random.random()*1.5+random.random()
else: else:
weights[attr] += delta weights[attr] += delta
if attr not in ['sigma', 'mu', 'se']: if attr not in ['sigma', 'mu', 'se']:
weights[attr] = min(max(0, weights[attr]), 3) weights[attr] = min(max(0.0, weights[attr]), 3.0)
mse = evaluateFitness(books) mse = evaluateFitness(books)
if mse < best_mse: # got better if mse < best_mse: # got better
saveWeights(weights) saveWeights(weights)
gamma = max(gamma*1.75, 0.001) gamma = min(max(gamma*1.75, 0.001), 0.5)
bestWeights = copy.copy(weights) bestWeights = copy.copy(weights)
best_mse = mse best_mse = mse
delta *= 2 delta *= 2