def _sampleWidthByAlphas(self):
# define Binomial distribution on n-1 layer filters (because we have to choose at least one filter)
dist = Binomial(self.outputChannels() - 1, logits=self._alphas)
# draw from distribution
width = 1 + dist.sample().type(int32).item()
return width
def calcNewWidthFunc(width: int,
alphaWidth: AlphaPerWidthBlock.AlphaWidth):
# define Binomial distribution on n-1 layer filters (because we have to choose at least one filter)
dist = Binomial(width - 1, logits=alphaWidth.tensor())
# draw from distribution
return 1 + dist.sample().type(int32).item()
def create_distribution(self, distribution_params):
return Binomial(1, logits=distribution_params)
# In[141]: dist = Beta(torch.tensor([0.5]), torch.tensor([0.5])) dist # In[142]: dist.sample() # In[143]: from torch.distributions.binomial import Binomial # In[144]: dist = Binomial(100, torch.tensor([0, .2, .8, 1])) # In[147]: dist.sample() # In[148]: # 100- count of trials # 0, 0.2, 0.8 and 1 are event probabilities # In[174]: from torch.distributions.categorical import Categorical # In[175]:
model_path = os.path.join('checkpoints', model_name)
if not os.path.exists(model_path):
os.makedirs(model_path)
from tensorboardX import SummaryWriter
writer = SummaryWriter(os.path.join('runs', model_name))
n_iter = 0
num_samples = 50
sample = torch.randn(64, opts.nz).double().to(device)
for epoch in range(opts.epochs):
print('=> Epoch {}'.format(epoch))
model.train()
running_loss = []
for data in tqdm(train_loader):
image = data[0]
m = Binomial(1, image.view(-1, 784))
# inputs = m.sample(torch.Size([num_samples])).double().to(device)
inputs = m.sample().expand(num_samples, image.shape[0],
784).double().to(device)
optimizer.zero_grad()
loss, bce, kld = model.train_loss(inputs)
loss.backward()
optimizer.step()
running_loss.append(loss.item())
writer.add_scalar('bce', bce, n_iter)
writer.add_scalar('kld', kld, n_iter)
writer.add_scalar('loss', loss, n_iter)
n_iter += 1
writer.add_scalar('loss_epoch', np.mean(running_loss), epoch)
def p_infected(self, n, p):
binom_dist = Binomial(n, p)
# returns 1 - p(0 transmissions to susceptible | 'n' nearby infected, 'p' prob. of transmission)
p0 = binom_dist.log_prob(torch.tensor(0.))
out = torch.log(1 - torch.exp(p0))
return out