def __init__(self, opt): """Initialize this dataset class. Parameters: opt (Option class) -- stores all the experiment flags; needs to be a subclass of BaseOptions """ BaseDataset.__init__(self, opt) # self.dir_A = os.path.join(opt.dataroot, opt.phase + 'B') # create a path '/path/to/data/trainA' self.dir_C = os.path.join(opt.dataroot, 'C/' + opt.phase) self.dir_E = os.path.join(opt.dataroot, 'E/' + opt.phase) self.dir_M = os.path.join(opt.dataroot, 'M/' + opt.phase) self.dir_B = os.path.join( opt.dataroot, 'B/' + opt.phase) # create a path '/path/to/data/trainB' # self.A_paths = sorted(make_dataset(self.dir_A, opt.max_dataset_size)) # load images from '/path/to/data/trainA' self.C_paths = sorted(make_dataset(self.dir_C, opt.max_dataset_size)) self.E_paths = sorted(make_dataset(self.dir_E, opt.max_dataset_size)) self.M_paths = sorted(make_dataset(self.dir_M, opt.max_dataset_size)) self.B_paths = sorted(make_dataset( self.dir_B, opt.max_dataset_size)) # load images from '/path/to/data/trainB' # self.A_size = len(self.A_paths) # get the size of dataset B self.C_size = len(self.C_paths) self.E_size = len(self.E_paths) self.M_size = len(self.M_paths) self.B_size = len(self.B_paths) # get the size of dataset ori btoA = self.opt.direction == 'BtoA' input_nc = self.opt.output_nc if btoA else self.opt.input_nc # get the number of channels of input image output_nc = self.opt.input_nc if btoA else self.opt.output_nc # get the number of channels of output image # self.transform_A = get_transform(self.opt, grayscale=(input_nc == 1)) self.transform_C = get_transform(self.opt, grayscale=False) self.transform_E = get_transform(self.opt, grayscale=True) self.transform_M = get_transform(self.opt, grayscale=True) self.transform_B = get_transform(self.opt, grayscale=(output_nc == 1))
def __init__(self, root, transform=None): super(ilab_sup_imgfolder, self).__init__() self.root = root self.transform = transform self.dir_Appe = os.path.join(self.root, 'appearance') self.Appe_paths = make_dataset(self.dir_Appe) self.Appe_size = len(self.Appe_paths)
def __init__(self, root, transform=None): super(ilab_threeswap_imgfolder, self).__init__() self.root = root self.transform = transform self.paths = make_dataset(self.root) self.C_size = len(self.paths) self.id_dict = {} self.bg_dict = {} self.pose_dict = {} self.id_cnt = 0 self.bg_cnt = 0 self.pose_cnt = 0 for roots, dirs, files in os.walk('/home2/ilab2M_pose/train_img_c00_10class'): for file in files: category = file.split('-')[0] id = file.split('-')[1] background = file.split('-')[2] pose = file.split('-')[3] + file.split('-')[4] if id not in self.id_dict: self.id_dict[id] = self.id_cnt self.id_cnt += 1 if background not in self.bg_dict: self.bg_dict[background] = self.bg_cnt self.bg_cnt += 1 if pose not in self.pose_dict: self.pose_dict[pose] = self.pose_cnt self.pose_cnt += 1
def __init__(self, root, transform=None): super(ilab_threeswap_imgfolder, self).__init__() self.root = root self.transform = transform self.idf_dict = {} self.exp_dict = {} self.posef_dict = {} self.idf_cnt = 0 self.exp_cnt = 0 self.posef_cnt = 0 for roots, dirs, files in os.walk('/home2/RaFD/train/data/'): for file in files: idf = file.split('_')[0] expression = file.split('_')[2].split('.')[0] posef = file.split('_')[1] if idf not in self.idf_dict: self.idf_dict[idf] = self.idf_cnt self.idf_cnt += 1 if expression not in self.exp_dict: self.exp_dict[expression] = self.exp_cnt self.exp_cnt += 1 if posef not in self.posef_dict: self.posef_dict[posef] = self.posef_cnt self.posef_cnt += 1 print(root) self.paths = make_dataset(self.root) file = open('debug.txt', 'w') file.write(str(self.paths)) file.close() self.C_size = len(self.paths) - 1 print(self.C_size)
def __init__(self, mode='Train'): super(SiameseDataset, self).__init__() self.mode = mode self.train_dir = './data/photo_parse_train/' self.test_dir = './data/photo_parse_test/' self.cari_dir = './data/caricature_parse/' self.train_path = make_dataset(self.train_dir) self.train_paths = sorted(self.train_path) self.train_dict = {} for img_name in self.train_paths: person_name = img_name.split('/')[-1] person_name = person_name[:-11] if person_name not in self.train_dict.keys(): self.train_dict[person_name] = [img_name] else: self.train_dict[person_name].append(img_name) self.train_size = len(self.train_path) self.test_path = make_dataset(self.test_dir) self.test_paths = sorted(self.test_path) self.test_size = len(self.test_path) self.cari_path = make_dataset(self.cari_dir) self.cari_paths = sorted(self.cari_path) self.cari_dict = {} for img_name in self.cari_paths: person_name = img_name.split('/')[-1] person_name = person_name[:-11] if person_name not in self.cari_dict.keys(): self.cari_dict[person_name] = [img_name] else: self.cari_dict[person_name].append(img_name) self.cari_size = len(self.cari_path) transform_list = [] transform_list += [transforms.ToTensor()] self.transform = transforms.Compose(transform_list) self.target_transform = self.transform if self.mode == 'Val': self.train_size = self.test_size self.train_path = self.test_path self.train_paths = self.test_paths
def __init__(self, opt): """Initialize this dataset class. Parameters: opt (Option class) -- stores all the experiment flags; needs to be a subclass of BaseOptions """ BaseDataset.__init__(self, opt) self.dir_AB = os.path.join(opt.dataroot, opt.phase) # get the image directory self.AB_paths = sorted(make_dataset(self.dir_AB, opt.max_dataset_size)) # get image paths assert(self.opt.load_size >= self.opt.crop_size) # crop_size should be smaller than the size of loaded image self.input_nc = self.opt.output_nc if self.opt.direction == 'BtoA' else self.opt.input_nc self.output_nc = self.opt.input_nc if self.opt.direction == 'BtoA' else self.opt.output_nc
def __init__(self, args): self.args = args self.data_path = args.data_path self.dir_AB = os.path.join(args.data_path, args.mode) # ??? self.AB_paths = sorted(make_dataset(self.dir_AB)) assert (agrs.resize_or_crop == 'resize_and_crop') transform_list = [ transforms.ToTensor(), transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5)) ] self.transform = transforms.Compose(transform_list)
def __init__(self, root, transform=None): super(ilab_unsup_imgfolder, self).__init__() self.root = root self.transform = transform self.paths = make_dataset(self.root) self.C_size = len(self.paths)
def __init__(self, dataroot, csv, batch_size=40): self.batchSize = batch_size self.root = dataroot self.frame = pd.read_csv(csv, header=None) self.image_paths = sorted(make_dataset(self.root)) self.dataset_size = len(self.image_paths)