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

Nicer prints when training

Browse Source
This commit is contained in:
Dominik Moritz Roth 2022-04-16 12:32:16 +02:00
parent cb84796d8f
commit 8353dc13d0
2 changed files with 15 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

BIN
brains/uttt.pth
View File

Binary file not shown.

28
vacuumDecay.py
View File

@ -473,23 +473,19 @@ class Trainer(Runtime):
self.rootNode = self.head self.rootNode = self.head
self.terminal = None self.terminal = None
def buildDatasetFromModel(self, model, depth=4, refining=True, exacity=5): def buildDatasetFromModel(self, model, depth=4, refining=True, fanOut=[16,16,8,8,6,6,5,4], uncertainSec=15, exacity=5):
print('[*] Building Timeline') print('[*] Building Timeline')
term = self.linearPlay(model, calcDepth=depth, exacity=exacity) term = self.linearPlay(model, calcDepth=depth, exacity=exacity)
if refining: if refining:
print('[*] Refining Timeline (exploring alternative endings)') print('[*] Refining Timeline (exploring alternative endings)')
self.fanOut(term, depth=depth+2) cur = term
self.fanOut(term.parent, depth=depth+2) for d in fanOut:
self.fanOut(term.parent.parent, depth=depth+2) cur = cur.parent
cur.forceStrong(d)
print('[*] Refining Timeline (exploring uncertain regions)') print('[*] Refining Timeline (exploring uncertain regions)')
self.timelineExpandUncertain(term, 10) self.timelineExpandUncertain(term, uncertainSec)
return term return term
def fanOut(self, head, depth=4):
for d in range(max(1, depth-2)):
head = head.parent
head.forceStrong(depth)
def linearPlay(self, model, calcDepth=7, exacity=5, verbose=True): def linearPlay(self, model, calcDepth=7, exacity=5, verbose=True):
head = self.rootNode head = self.rootNode
self.universe.model = model self.universe.model = model
@ -510,7 +506,9 @@ class Trainer(Runtime):
else: else:
ind = int(pow(random.random(),exacity)*(len(opts)-1)) ind = int(pow(random.random(),exacity)*(len(opts)-1))
head = opts[ind][0] head = opts[ind][0]
print('') if verbose:
print(head)
print(' => '+['O','X','No one '][self.head.getWinner()] + ' won!')
return head return head
def timelineIter(self, term): def timelineIter(self, term):
@ -546,6 +544,8 @@ class Trainer(Runtime):
inp = node.state.getTensor(player=p) inp = node.state.getTensor(player=p)
gol = torch.tensor(node.getStrongFor(p), dtype=torch.float) gol = torch.tensor(node.getStrongFor(p), dtype=torch.float)
out = model(inp) out = model(inp)
if not out:
continue
loss = loss_func(out, gol) loss = loss_func(out, gol)
optimizer.zero_grad() optimizer.zero_grad()
loss.backward() loss.backward()
@ -557,8 +557,9 @@ class Trainer(Runtime):
break break
#print(loss_sum/i) #print(loss_sum/i)
if r > 16 and (loss_sum/i < cut or lLoss == loss_sum): if r > 16 and (loss_sum/i < cut or lLoss == loss_sum):
return return loss_sum
lLoss = loss_sum lLoss = loss_sum
return loss_sum
def main(self, model=None, gens=1024, startGen=12): def main(self, model=None, gens=1024, startGen=12):
newModel = False newModel = False
@ -568,7 +569,8 @@ class Trainer(Runtime):
self.universe.scoreProvider = ['neural','naive'][newModel] self.universe.scoreProvider = ['neural','naive'][newModel]
for gen in range(startGen, startGen+gens): for gen in range(startGen, startGen+gens):
print('[#####] Gen '+str(gen)+' training:') print('[#####] Gen '+str(gen)+' training:')
self.trainModel(model, calcDepth=min(5,3+int(gen/16)), exacity=int(gen/3+1)) loss = self.trainModel(model, calcDepth=min(5,3+int(gen/16)), exacity=int(gen/3+1))
print('[L] '+str(loss))
self.universe.scoreProvider = 'neural' self.universe.scoreProvider = 'neural'
self.saveModel(model) self.saveModel(model)