classify_img_sliced: Classify images as superpixel slices

In jspaezp/clasifierrr: Classifies images based on features

Description

Classify images as superpixel slices

Usage

classify_img_sliced(
  classifier,
  img,
  slices = slic(sobel_filter(img, 5))[[2]],
  features,
  class_highlight,
  ...
)

Arguments

classifier	a classifier
img	image to classify
features	a features frame to be used for the classification
class_highlight	a classiication to highlight if the prediction is categorical
...	additional arguments passed to kmeans

See Also

slic

Examples

# trainset <- build_train_multi(
# params_df,
# train_size_each = 5000,
# filter_widths = c(3,5))
#
# classifier <- ranger::ranger(pixel_class ~ ., trainset)
# features <- calc_features(base_img, c(3,5))
# sliced_img <- slic(sobel_filter(base_img, 5))
# classify_img_sliced(classifier, base_img, sliced_img[[2]],features, "sphere")

jspaezp/clasifierrr documentation built on March 2, 2020, 11:20 a.m.