extract=False,
dropout_p=args.dropout_p).cuda()
netG.eval()
netF.eval()
for epoch in range(args.epoch / 2, args.epoch + 1):
if epoch % 10 != 0:
continue
netG.load_state_dict(
torch.load(
os.path.join(
args.snapshot, "Office31_HAFN_" + args.task + "_netG_" +
args.post + "." + args.repeat + "_" + str(epoch) + ".pth")))
netF.load_state_dict(
torch.load(
os.path.join(
args.snapshot, "Office31_HAFN_" + args.task + "_netF_" +
args.post + "." + args.repeat + "_" + str(epoch) + ".pth")))
correct = 0
tick = 0
for (imgs, labels) in t_loader:
tick += 1
imgs = Variable(imgs.cuda())
pred = netF(netG(imgs))
pred = F.softmax(pred)
pred = pred.data.cpu().numpy()
pred = pred.argmax(axis=1)
labels = labels.numpy()
correct += np.equal(labels, pred).sum()
correct = correct * 1.0 / len(t_set)
netF = ResClassifier(class_num=args.class_num).cuda()
netG.eval()
netF.eval()
for epoch in range(args.epoch / 2, args.epoch + 1):
if epoch % 10 != 0:
continue
netG.load_state_dict(
torch.load(
os.path.join(
args.snapshot, "VisDA_IAFN_netG_" + args.post + '.' +
str(args.repeat) + '_' + str(epoch) + ".pth")))
netF.load_state_dict(
torch.load(
os.path.join(
args.snapshot, "VisDA_IAFN_netF_" + args.post + '.' +
str(args.repeat) + '_' + str(epoch) + ".pth")))
correct = 0
tick = 0
for (imgs, labels) in t_loader:
tick += 1
imgs = Variable(imgs.cuda())
pred = netF(netG(imgs))
pred = F.softmax(pred)
pred = pred.data.cpu().numpy()
pred = pred.argmax(axis=1)
labels = labels.numpy()
correct += np.equal(labels, pred).sum()
netG = ResBase50().cuda()
netF = ResClassifier(class_num=args.class_num, extract=False).cuda()
netG.eval()
netF.eval()
for epoch in range(1, args.epoch + 1):
if epoch % 10 != 0:
continue
netG.load_state_dict(
torch.load(
os.path.join(
args.snapshot, "OfficeHome_IAFN_" + args.task + "_netG_" +
args.post + '.' + args.repeat + '_' + str(epoch) + ".pth")))
netF.load_state_dict(
torch.load(
os.path.join(
args.snapshot, "OfficeHome_IAFN_" + args.task + "_netF_" +
args.post + '.' + args.repeat + '_' + str(epoch) + ".pth")))
correct = 0
tick = 0
for (imgs, labels) in t_loader:
tick += 1
imgs = Variable(imgs.cuda())
pred = netF(netG(imgs))
pred = F.softmax(pred)
pred = pred.data.cpu().numpy()
pred = pred.argmax(axis=1)
labels = labels.numpy()
correct += np.equal(labels, pred).sum()