clara/pytorch-ai
1
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

Minor Fixes for GPU training

Browse source
This commit is contained in:
Clemens-Dautermann 2020-01-28 15:23:25 +01:00
parent 56ee2635b5
commit 459caccfe2
47 changed files with 9862 additions and 26 deletions
Download patch file Download diff file Expand all files Collapse all files

18
TicTacToe_AI/Net/pytorch_ai.py
View file

@ -8,7 +8,7 @@ import wandb
wandb.init(project="tictactoe")
BATCH_SIZE = 3
BATCH_SIZE = 250
def to_set(raw_list):
@ -85,17 +85,19 @@ def buildsets():
testset = to_batched_set(alllines[0:10000])
print('Generating trainset...')
trainset = to_batched_set(alllines[10001:20000])
trainset = to_batched_set(alllines[10001:])
return trainset, testset
def testnet(net, testset):
def testnet(net, testset, device):
correct = 0
total = 0
with torch.no_grad():
for X, label in testset:
X = X.to(device)
output = net(X)
output = output.cpu()
if torch.argmax(output) == label[0]:
correct += 1
total += 1
@ -134,18 +136,20 @@ loss_function = nn.CrossEntropyLoss()
trainset, testset = buildsets()
for epoch in range(100):
for epoch in range(300):
print('Epoch: ' + str(epoch))
wandb.log({'epoch': epoch})
for X, label in tqdm(trainset):
net.zero_grad()
X.to(device)
X = X.to(device)
output = net(X)
output.cpu()
output = output.cpu()
loss = loss_function(output.view(-1, 10), label)
loss.backward()
optimizer.step()
wandb.log({'loss': loss})
net = net.cpu()
torch.save(net, './nets/gpunets/net_' + str(epoch) + '.pt')
testnet(net, testset)
net = net.to(device)
testnet(net, testset, device)