Transforms

Image transformation functions

transform_adjust_brightness()

Adjust the brightness of an image

transform_adjust_contrast()

Adjust the contrast of an image

transform_adjust_gamma()

Adjust the gamma of an RGB image

transform_adjust_hue()

Adjust the hue of an image

transform_adjust_saturation()

Adjust the color saturation of an image

transform_affine()

Apply affine transformation on an image keeping image center invariant

transform_center_crop()

Crops the given image at the center

transform_color_jitter()

Randomly change the brightness, contrast and saturation of an image

transform_convert_image_dtype()

Convert a tensor image to the given dtype and scale the values accordingly

transform_crop()

Crop the given image at specified location and output size

transform_five_crop()

Crop image into four corners and a central crop

transform_grayscale()

Convert image to grayscale

transform_hflip()

Horizontally flip a PIL Image or Tensor

transform_linear_transformation()

Transform a tensor image with a square transformation matrix and a mean_vector computed offline

transform_normalize()

Normalize a tensor image with mean and standard deviation

transform_pad()

Pad the given image on all sides with the given "pad" value

transform_perspective()

Perspective transformation of an image

transform_random_affine()

Random affine transformation of the image keeping center invariant

transform_random_apply()

Apply a list of transformations randomly with a given probability

transform_random_choice()

Apply single transformation randomly picked from a list

transform_random_crop()

Crop the given image at a random location

transform_random_erasing()

Randomly selects a rectangular region in an image and erases its pixel values

transform_random_grayscale()

Randomly convert image to grayscale with a given probability

transform_random_horizontal_flip()

Horizontally flip an image randomly with a given probability

transform_random_order()

Apply a list of transformations in a random order

transform_random_perspective()

Random perspective transformation of an image with a given probability

transform_random_resized_crop()

Crop image to random size and aspect ratio

transform_random_rotation()

Rotate the image by angle

transform_random_vertical_flip()

Vertically flip an image randomly with a given probability

transform_resize()

Resize the input image to the given size

transform_resized_crop()

Crop an image and resize it to a desired size

transform_rgb_to_grayscale()

Convert RGB Image Tensor to Grayscale

transform_rotate()

Angular rotation of an image

transform_ten_crop()

Crop an image and the flipped image each into four corners and a central crop

transform_to_tensor()

Convert an image to a tensor

transform_vflip()

Vertically flip a PIL Image or Tensor

Models

Model architectures

model_alexnet()

AlexNet Model Architecture

model_inception_v3()

Inception v3 model

model_mobilenet_v2()

Constructs a MobileNetV2 architecture from MobileNetV2: Inverted Residuals and Linear Bottlenecks.

model_resnet18() model_resnet34() model_resnet50() model_resnet101() model_resnet152() model_resnext50_32x4d() model_resnext101_32x8d() model_wide_resnet50_2() model_wide_resnet101_2()

ResNet implementation

model_vgg11() model_vgg11_bn() model_vgg13() model_vgg13_bn() model_vgg16() model_vgg16_bn() model_vgg19() model_vgg19_bn()

VGG implementation

Datasets

Datasets readily available

cifar10_dataset() cifar100_dataset()

Cifar datasets

image_folder_dataset()

Create an image folder dataset

kmnist_dataset()

Kuzushiji-MNIST

mnist_dataset()

MNIST dataset

tiny_imagenet_dataset()

Tiny ImageNet dataset

Displaying

Show images

draw_bounding_boxes()

Draws bounding boxes on image.

draw_keypoints()

Draws Keypoints

draw_segmentation_masks()

Draw segmentation masks

tensor_image_browse()

Display image tensor

tensor_image_display()

Display image tensor

Misc

magick_loader()

Load an Image using ImageMagick

base_loader()

Base loader

vision_make_grid()

A simplified version of torchvision.utils.make_grid