print('---------- Networks architecture -------------')
print_network(model)
print('----------------------------------------------')
if opt.resume:
if os.path.isfile(opt.resume):
print("=> loading checkpoint '{}'".format(opt.resume))
checkpoint = torch.load(opt.resume)
opt.start_epoch = checkpoint["epoch"] + 1
print(opt.start_epoch)
model.load_state_dict(checkpoint["model"].state_dict())
else:
print("=> no checkpoint found at '{}'".format(opt.resume))
if cuda:
model = model.cuda()
criterion = criterion.cuda()
optimizer = optim.Adam(model.parameters(),
lr=opt.lr,
betas=(0.9, 0.999),
eps=1e-8)
for epoch in range(opt.start_epoch, opt.nEpochs + 1):
train(model, epoch)
# learning rate is decayed by a factor of 2 every 200 epochs
if (epoch + 1) % 500 == 0:
for param_group in optimizer.param_groups:
param_group['lr'] /= 10.0
print('Learning rate decay: lr={}'.format(
features_test = np.load(
os.path.join(config["load_dir"] + '/_feature_test_assigment.npy'))
features_test_crop = np.load(
os.path.join(config["load_dir"] + '/_feature_test_assigment_crop.npy'))
if config["concatenate"]:
features_test = np.concatenate((features_test, features_test_crop), axis=1)
#Model
model = Net(features_test.shape[1])
model.load_state_dict(state_dict['model'])
model.eval()
if use_cuda:
print('Using GPU')
model.cuda()
else:
print('Using CPU')
for dossier, sous_dossiers, fichiers in os.walk(config["test_dir"]):
for num, fichier in enumerate(fichiers):
num_photo = df.loc[df['Id'] == fichier.split('.')[0]].index[0]
data = torch.tensor(features_test[num])
if use_cuda:
data = data.cuda()
output = model(data)
prout, pred = torch.max(output.data, 0)
df.Category[num_photo] = pred
df.to_csv('kaggle.csv', index=False)
print(
