def joint_transform(image, mask, weight):
# transforming to PIL image
image, mask, weight = F.to_pil_image(image), F.to_pil_image(
mask), F.to_pil_image(weight)
# random crop
i, j, h, w = T.RandomCrop.get_params(image, size)
image, mask, weight = F.crop(image, i, j, h,
w), F.crop(mask, i, j, h,
w), F.crop(weight, i, j, h, w)
if np.random.rand() < p_flip:
image, mask, weight = F.hflip(image), F.hflip(mask), F.hflip(
weight)
# color transforms || ONLY ON IMAGE
if color_tf is not None:
image = color_tf(image)
# transforming to tensor
image, weight = F.to_tensor(image), F.to_tensor(weight)
if not long_mask:
mask = F.to_tensor(mask)
else:
mask = to_long_tensor(mask)
return image, mask, weight
def load_image(fn, size, path, aug, y):
img2 = cv2.imread(path + '/' + fn)
img = cv2.resize(img2, (size, size))
# print(img.shape)
if aug == 0:
yy = y
if aug == 1:
#rotate 90 clock
img = cv2.transpose(img)
img = cv2.flip(img, 1)
yy = [y[3], y[0], y[1], y[2]]
elif aug == 2:
#rotate 90 anticlock
img = cv2.transpose(img)
img = cv2.flip(img, 1)
yy = [y[1], y[3], y[2], y[0]]
elif aug == 3:
# flip v
img = cv2.flip(img, 0)
yy = [y[0], y[3], y[2], y[1]]
elif aug == 4:
# flip h
img = cv2.flip(img, 1)
yy = [y[2], y[1], y[0], y[3]]
img = F.to_tensor(img)
yy = torch.tensor(np.array(yy), dtype=torch.float)
return img, yy
def joint_transform(image, mask):
# random magnification
if random_resize:
magnification_ratio = (random_resize[1] - random_resize[0]
) * np.random.rand() + random_resize[0]
new_shape = (int(magnification_ratio * image.shape[0]),
int(magnification_ratio * image.shape[1]))
image = img_as_ubyte(resize(image, new_shape))
mask = resize(mask, new_shape)
mask = img_as_ubyte(mask > 0.5)
# resizing
if image.shape[0] < size[0] or image.shape[1] < size[1]:
new_im_shape = np.max([image.shape[0],
size[0]]), np.max([image.shape[1],
size[1]]), 3
new_mask_shape = np.max([image.shape[0], size[0]
]), np.max([image.shape[1], size[1]]), 1
image = pad_to_shape(image, new_im_shape)
mask = pad_to_shape(mask, new_mask_shape)
# transforming to PIL image
image, mask = F.to_pil_image(image), F.to_pil_image(mask)
# random crop
i, j, h, w = T.RandomCrop.get_params(image, size)
image, mask = F.crop(image, i, j, h, w), F.crop(mask, i, j, h, w)
if np.random.rand() < p_flip:
image, mask = F.hflip(image), F.hflip(mask)
# color transforms || ONLY ON IMAGE
if color_tf is not None:
image = color_tf(image)
# transforming to tensor
image = F.to_tensor(image)
if not long_mask:
mask = F.to_tensor(mask)
else:
mask = to_long_tensor(mask)
# normalizing image
if normalize:
image = tf_normalize(image)
return image, mask
def __call__(self, pic):
"""
Args:
pic (PIL Image or numpy.ndarray): Image to be converted to tensor.
Returns:
Tensor: Converted image.
"""
return F.to_tensor(pic)
def transform(image):
# transforming to PIL image
image = F.to_pil_image(image)
# random crop
i, j, h, w = T.RandomCrop.get_params(image, size)
image = F.crop(image, i, j, h, w)
if np.random.rand() < p_flip:
image = F.hflip(image)
# color transforms || ONLY ON IMAGE
if color_tf is not None:
image = color_tf(image)
# transforming to tensor
image = F.to_tensor(image)
return image
def single_transform(image):
# resizing
if image.shape[0] < size[0] or image.shape[1] < size[1]:
new_im_shape = np.max([image.shape[0],
size[0]]), np.max([image.shape[1],
size[1]]), 3
image = resize(image, new_im_shape, preserve_range=True).astype(
np.uint8) #pad_to_shape(image, new_im_shape)
# transforming to PIL image
image = F.to_pil_image(image)
# transforming to tensor
image = F.to_tensor(image)
# normalizing image
if normalize:
image = tf_normalize(image)
return image
def joint_transform(image, mask, rcnn_mask):
# resizing
if image.shape[0] < size[0] or image.shape[1] < size[1]:
new_im_shape = np.max([image.shape[0],
size[0]]), np.max([image.shape[1],
size[1]]), 3
new_mask_shape = np.max([image.shape[0], size[0]
]), np.max([image.shape[1], size[1]]), 1
image = pad_to_shape(image, new_im_shape)
mask, rcnn_mask = pad_to_shape(mask, new_mask_shape), pad_to_shape(
rcnn_mask, new_mask_shape)
# transforming to PIL image
image, mask, rcnn_mask = list(
map(F.to_pil_image, [image, mask, rcnn_mask]))
# random crop
i, j, h, w = T.RandomCrop.get_params(image, size)
image, mask, rcnn_mask = F.crop(image, i, j, h, w), F.crop(
mask, i, j, h, w), F.crop(rcnn_mask, i, j, h, w)
# random flip
if np.random.rand() < p_flip:
image, mask, rcnn_mask = list(
map(F.hflip, [image, mask, rcnn_mask]))
if np.random.rand() < p_flip:
image, mask, rcnn_mask = list(
map(F.vflip, [image, mask, rcnn_mask]))
# transforming to tensor
image, rcnn_mask = list(map(F.to_tensor, [image, rcnn_mask]))
if not long_mask:
mask = F.to_tensor(mask)
else:
mask = to_long_tensor(mask)
return image, mask, rcnn_mask
def __call__(self, image, target):
image = F.to_tensor(image)
return image, target
def __call__(self, image, target):
image = F.to_tensor(image) # bbox坐标、label原本就是Tensor，无需转换
return image, target
def __call__(self, sample):
image, labels = sample['image'], sample['labels']
image = F.to_tensor(image)
return {'image': image, 'labels': labels}
train=True,
download=True,
transform=transforms.ToTensor())
mnist_train = DataLoader(mnist_train, batch_size=32)
mnist_val = MNIST(os.getcwd(),
train=True,
download=True,
transform=transforms.ToTensor())
mnist_val = DataLoader(mnist_val, batch_size=11264)
# train_loader, val_loader = get_data_loaders(1024, 8192)
model = Net()
model.cuda()
model.layer = nn.DataParallel(model.layer)
# most basic trainer, uses good defaults
trainer = Trainer(progress_bar_refresh_rate=0.4, max_epochs=30, gpus=1)
trainer.fit(model, mnist_train, mnist_val)
import matplotlib.pyplot as plt
img, lbl = next(iter(mnist_val))
for i in img:
preds = model(i.unsqueeze(0).cuda())
plt.title(preds.argmax())
plt.imshow(i.permute((1, 2, 0)))
plt.show()
import torchvision.transforms.functional as F
F.to_tensor()
def __call__(self, pic, image_name):
return F.to_tensor(pic)
