ansformm = transforms.Lambda(lambda x: TF.affine(x, Here is a simple example for a folder containing 5 images: class CustomDataset(Dataset):ĭef _init_(self, image_paths, transform_images): In _getitem_ you should compare the current image_path not the image itself. Try to pass a list containing all images names which should be transformed. I also don’t have any idea how can I do it for mask of images (my task is segmentation) at the moment is giving an error _init_() takes 3 positional arguments but 4 were given I really appreciate if you point me to the right direction. In the second snippet transform_images = list(folder_data) is list of image name that I need to augment. Test_loader = (test_dataset, batch_size=1, shuffle=False, num_workers=2) Test_dataset = CustomDataset(test_image_paths, test_mask_paths, transform_images) Valid_loader = (valid_dataset, batch_size=1, shuffle=True, num_workers=2) Valid_dataset = CustomDataset(valid_image_paths, valid_mask_paths, transform_images) Train_loader = (train_dataset, batch_size=1, shuffle=True, num_workers=2) Train_dataset = CustomDataset(train_image_paths, train_mask_paths, transform_images) Transform_images = glob.glob("F:\\my_data\\imagesResized\\P164_ES_1.png") Print("count of validation mask is: ", len(valid_image_paths)) Print("count of validation image is: ", len(valid_image_paths)) Print("count of training mask is: ", len(train_image_paths)) Print("count of training images is: ", len(train_image_paths))
Print("count of test mask is: ", len(test_mask_paths))Īssert len(folder_data) = len(folder_mask) Print("count of test images is: ", len(test_image_paths)) Return t_image, mask, self.image_paths, self.target_pathsĪnd here is there snippet for splitting the dataset and define dataloaders from custom_dataset import CustomDatasetįolder_data = glob.glob("F:\\my_data\\imagesResized\\*.png")įolder_mask = glob.glob("F:\\my_data\\labelsResized\\*.png") Mask = om_numpy(numpy.array(mask, dtype=numpy.uint8)) T_image = ansformm(t_image) #augmentation T_image = ansforms(t_image) # transform to tensor ansformm = transforms.Compose([tf.rotate(10),
I have been trying to do augmentation with no luck.ĭef _init_(self, image_paths, target_paths, transform_images):
0 kommentar(er)
