Class RandomResizedCrop

Inheritance Relationships

Base Type

Class Documentation

class RandomResizedCrop : public mindspore::dataset::TensorTransform

Crop the input image to a random size and aspect ratio.

Public Functions

explicit RandomResizedCrop(const std::vector<int32_t> &size, const std::vector<float> &scale = {0.08, 1.0}, const std::vector<float> &ratio = {3. / 4., 4. / 3.}, InterpolationMode interpolation = InterpolationMode::kLinear, int32_t max_attempts = 10)

Constructor.

Example
/* Define operations */
auto decode_op = vision::Decode();
auto random_op = vision::RandomResizedCrop({32, 32}, {0.08, 1.0});

/* dataset is an instance of Dataset object */
dataset = dataset->Map({decode_op, random_op},  // operations
                       {"image"});              // input columns

说明

If the input image is more than one, then make sure that the image size is the same.

参数
  • size[in] A vector representing the output size of the cropped image. If the size is a single value, a squared crop of size (size, size) is returned. If the size has 2 values, it should be (height, width).

  • scale[in] Range [min, max) of respective size of the original size to be cropped (default=(0.08, 1.0)).

  • ratio[in] Range [min, max) of aspect ratio to be cropped (default=(3. / 4., 4. / 3.)).

  • interpolation[in] Image interpolation mode (default=InterpolationMode::kLinear).

    • InterpolationMode::kLinear, Interpolation method is blinear interpolation.

    • InterpolationMode::kNearestNeighbour, Interpolation method is nearest-neighbor interpolation.

    • InterpolationMode::kCubic, Interpolation method is bicubic interpolation.

    • InterpolationMode::kArea, Interpolation method is pixel area interpolation.

    • InterpolationMode::kCubicPil, Interpolation method is bicubic interpolation like implemented in pillow.

  • max_attempts[in] The maximum number of attempts to propose a valid. crop_area (default=10). If exceeded, fall back to use center_crop instead.

~RandomResizedCrop() = default

Destructor.