def predict():
model = Net(model_name).to(device)
model_save_path = os.path.join(config.model_path,
'{}.bin'.format(model_name))
model.load_state_dict(torch.load(model_save_path))
data_len = len(os.listdir(config.image_test_path))
test_path_list = [
'{}/{}.jpg'.format(config.image_test_path, x)
for x in range(0, data_len)
]
test_data = np.array(test_path_list)
test_dataset = MyDataset(test_data, test_transform, 'test')
test_loader = DataLoader(test_dataset,
batch_size=config.batch_size,
shuffle=False)
model.eval()
pred_list = []
with torch.no_grad():
for batch_x, _ in tqdm(test_loader):
batch_x = batch_x.to(device)
# compute output
logits = model(batch_x)
preds = torch.argmax(logits, dim=1)
pred_list += [p.cpu().data.numpy() for p in preds]
submission = pd.DataFrame({
"id": range(len(pred_list)),
"label": pred_list
})
submission.to_csv('submission.csv', index=False, header=False)
def test(**kwargs):
config.parse(kwargs)
test_loader = DataLoader(
WholeFaceDatasets(config.whole_csv_test, config.whole_imgs_base_dir, train=False, transform=transforms.Compose([
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])), batch_size=config.batch_size, shuffle=False,
num_workers=config.num_workers
)
model = Net()
if config.using_pretrain_model:
model.load_state_dict(torch.load(config.pretrain_model_path))
model.cuda()
model.eval()
test_loss = 0
tables = None
for data, target in test_loader:
target = target.type(torch.FloatTensor)
data, target = data.cuda(), target.cuda()
data, target = Variable(data, volatile=True), Variable(target)
output = model(data)
# print(output)
loss = F.l1_loss(output, target).cpu()
a = output.cpu().data.numpy()
b = np.reshape(target.cpu().data.numpy(), (-1, 1))
pair = np.hstack((a, b))
if tables is None:
tables = pair
else:
tables = np.vstack((tables, pair))
test_loss += loss.data[0] * config.batch_size
test_loss /= len(test_loader.dataset)
# print(tables)
np.save('predicted_and_real.npy', tables)
print(f'Testing Accuracy is {test_loss}')