def test(model_config):
mode = 'test'
batch_size = 1
dataset = ShakespeareModern(train_shakespeare_path, test_shakespeare_path, train_modern_path, test_modern_path, mode=mode)
dataloader = DataLoader(dataset, batch_size=batch_size, shuffle=False)
shakespeare_disc = Discriminator(model_config['embedding_size'], model_config['hidden_dim'], len(dataset.vocab)).cuda()
shakespeare_disc.load_state_dict(torch.load('./shakespeare_disc.pth'))
shakespeare_disc.eval()
num_correct = 0
total_samples = 0
for idx, (s, s_addn_feats, m, m_addn_feats) in tqdm(enumerate(dataloader)):
s = s.transpose(0, 1)
m = m.transpose(0, 1)
total_samples += 2
s = Variable(s).cuda()
s_output = shakespeare_disc(s, s_addn_feats)
if round(s_output.item()) == 1.0:
num_correct += 1
m = Variable(m).cuda()
m_output = shakespeare_disc(m, m_addn_feats)
if round(m_output.item()) == 0.0:
num_correct += 1
print ('Accuracy: {}'.format(num_correct/total_samples))
fake_norm = fake.data[0].norm()
criticD_fake = netD(fake.detach(), batch_att)
criticD_fake = criticD_fake.mean()
criticD_fake.backward()
# gradient penalty loss
gradient_penalty = calc_gradient_penalty(netD, batch_feat, fake.data, batch_att, opts)
gradient_penalty.backward()
Wasserstein_D = criticD_real - criticD_fake
D_cost = criticD_fake - criticD_real + gradient_penalty
optimizerD.step()
netG.train()
netD.eval()
# Train G
netG.zero_grad()
noise.normal_(0, 1)
fake = netG(noise, batch_att)
criticG_fake = netD(fake, batch_att)
criticG_fake = criticG_fake.mean()
G_cost = -criticG_fake
# Classifier loss
c_errG = cls_criterion(pre_cla(fake), Variable(utils.map_label(batch_l, seenclasses)))
# Bi-sematic reconstruct loss
