def __init__(self):
self._opt = TrainOptions().parse()
self._dataset_train = DatasetFactory.get_by_name("FinetuneDataset", self._opt)
self._dataset_train_size = len(self._dataset_train)
print('#train images = %d' % self._dataset_train_size)
self._model = ModelsFactory.get_by_name("FineModel", self._opt, is_train=True)
self._train()
def __init__(self):
self._opt = TestOptions().parse()
self._img_path = self._opt.img_path
self._img_size = self._opt.image_size
self.fine_model = ModelsFactory.get_by_name('FineModel',
self._opt,
is_train=False)
self.svm_model_A = ModelsFactory.get_by_name('SvmModel',
self._opt,
is_train=False)
self.svm_model_A.load('A')
self.svm_model_B = ModelsFactory.get_by_name('SvmModel',
self._opt,
is_train=False)
self.svm_model_B.load('B')
self.svms = [self.svm_model_A, self.svm_model_B]
self.reg_model = ModelsFactory.get_by_name('RegModel',
self._opt,
is_train=False)
self.test()
def __init__(self):
self._opt = TrainOptions().parse()
self._dataset_train = DatasetFactory.get_by_name(
"SVMDataset", self._opt)
self._dataset_train_size = len(self._dataset_train)
print('#train images = %d' % self._dataset_train_size)
self.classA_features, self.classA_labels, self.classB_features, self.classB_labels = self._dataset_train.get_datas(
)
self._modelA = ModelsFactory.get_by_name("SvmModel",
self._opt,
is_train=True)
self._modelB = ModelsFactory.get_by_name("SvmModel",
self._opt,
is_train=True)
self._train(self._modelA, self.classA_features, self.classA_labels,
"A")
self._train(self._modelB, self.classB_features, self.classB_labels,
"B")
def __init__(self):
self._opt = TestOptions().parse()
self._img_path = self._opt.img_path
self._img_width = self._opt.image_width
self._img_height = self._opt.image_height
self._model = ModelsFactory.get_by_name('AlexModel',
self._opt,
is_train=False)
self._classes = self._opt.classes.split(",")
self._class_to_ind = dict(
zip(range(1,
len(self._classes) + 1), self._classes))
self._class_to_ind[0] = 'None'
self.test()
def __init__(self):
self._opt = TrainOptions().parse()
data_loader_train = CustomDatasetDataLoader(self._opt,
is_for_train=True)
#data_loader_test = CustomDatasetDataLoader(self._opt, is_for_train=False)
self._dataset_train = data_loader_train.load_data()
#self._dataset_test = data_loader_test.load_data()
self._dataset_train_size = len(data_loader_train)
#self._dataset_test_size = len(data_loader_test)
print('#train images = %d' % self._dataset_train_size)
#print('#test images = %d' % self._dataset_test_size)
self._model = ModelsFactory.get_by_name(self._opt.model, self._opt)
self._tb_visualizer = TBVisualizer(self._opt)
self._train()
def __init__(self, opt):
super(SVMDataset, self).__init__(opt)
self._name = 'SVMDataset'
self.datas = []
self.classA_features = []
self.classA_labels = []
self.classB_features = []
self.classB_labels = []
self.save_path = os.path.join(self._opt.generate_save_path, 'svm_data.npy')
self._img_size = self._opt.image_size
self.cursor = 0
self.model = ModelsFactory.get_by_name('FineModel', self._opt, is_train=False)
# read dataset
if os.path.exists(self.save_path):
self._load_from_numpy()
else:
self._load_dataset()