def __getitem__(self, index):
dir_path_name = self.dir_filenames[index]
path_list = os.listdir(dir_path_name)
path_list.sort()
name = path_list[0].split('.')[0]
path_csv = join(dir_path_name, path_list[0])
pd_data = pd.read_csv(path_csv).Exposure
label = np.array(pd_data)
label = torch.from_numpy(label).float()
path_list = [
join(dir_path_name, x) for x in path_list if is_image_file(x)
]
data_init = Image.open(path_list[0])
data_init = data_init.resize((1024, 1024))
if self.transform:
data_init = self.transform(data_init)
name = int(name)
for i in range(len(path_list) - 1):
data = Image.open(path_list[i + 1])
data = data.resize((1024, 1024))
if self.transform:
data = self.transform(data)
data_init = torch.cat((data_init, data), 0)
return data_init, label, name
def __getitem__(self, index):
dir_path_name = self.dir_filenames[index]
path_list = os.listdir(dir_path_name)
path_list.sort()
name = path_list[0].split('.')[0].split('_')[1]
path_list = [
join(dir_path_name, x) for x in path_list if is_image_file(x)
]
data_init = Image.open(path_list[0])
data_init = data_init.resize((4032, 3024))
if self.transform:
data_init = self.transform(data_init)
name = int(name)
for i in range(len(path_list) - 1):
data = Image.open(path_list[i + 1])
data = data.resize((4032, 3024))
if self.transform:
data = self.transform(data)
data_init = torch.cat((data_init, data), 0)
return data_init, name
def __getitem__(self, index):
sample = self.imgs[index]
splits = sample.split()
img_path = splits[0]
data = Image.open(img_path)
data = data.resize((224, 224))
#data = data.convert('L')
data = self.transforms(data)
label = np.int32(splits[1])
return data.float(), label
def __getitem__(self, index):
img_name = self.fileList[index]
data = PIL.Image.open('%s/%s' % (self.path, img_name))
height = 32
width = int(data.size[0] / (data.size[1] / height))
data = data.resize((width, height))
if self.transform is not None:
data = self.transform(data)
label = re.search('([0-9_]+)', img_name).group(1)
label = re.sub('\D', '', label)
return data, label
def img_open(path):
data = PIL.Image.open(path)
height = 32
width = int(data.size[0] / (data.size[1] / height))
data = data.resize((width, height))
Transform = transforms.Compose([
transforms.Grayscale(),
transforms.ToTensor(),
transforms.Lambda(lambda x: torch.unsqueeze(x, 0))
])
data = Transform(data)
return data
def __getitem__(self, index):
data_path = self.seq_list[index]
imgs = []
for img in data_path:
data = Image.open(img)
data = data.resize(self.size, Image.ANTIALIAS)
data = (np.asarray(data) / 255.0)
imgs.append(torch.from_numpy(data).float())
imgs = torch.stack(imgs)
return imgs.permute(0, 3, 1, 2)
def __getitem__(self, index):
sample = self.imgs[index]
race_list = ['African', 'Caucasian', 'Asian', 'Indian']
splits = sample.split()
img_path = splits[0]
race = race_list.index(img_path.split('/')[-3])
data = Image.open(img_path)
data = np.array(data.resize((128, 128)))
#data = data.convert('L')
data = self.transforms(data)
data = np.transpose(data, (2, 0, 1))
label = np.int32(race)
return torch.from_numpy(data).float(), label
def __getitem__(self, index):
img_path = self.train_data[index][0]
i = self.train_data[index][1]
id = self.train_data[index][2]
cam = self.train_data[index][3]
label = np.asarray(self.train_attr[id])
data = Image.open(img_path)
data = data.resize((64, 64))
#data=self.validate_image(data)
#data = self.transforms(data)
data = np.array(data, dtype=float)
data = torch.FloatTensor(data)
data = data.sub_(127.5).div_(127.5)
name = self.train_data[index][4]
return data, i, label, id, cam, name
def get_cla_sample(job):
image, x, y, z, dx, dy, size = job
if max(dx,dy) <=size+1:
r = size / 2
data = image[y - r:y + r, x - r:x + r, z - r:z + r].copy()#涉及浅拷贝问题:无法修改
if data.shape != (size, size, size):
data.resize((size,size,size))
else:
r = max(dx,dy)/2
data = image[y - r:y + r, x - r:x + r, z - r:z + r].copy()
if data.shape != (2*r,2*r,2*r):
data.resize((2*r,2*r,2*r))
data.resize((size,size,size))
return data