score_slice_data_dir: Score a directory of tile data files with a Keras model

In treysp/coolit: Identify water cooling towers from aerial orthoimagery

Description

Score a directory of tile data files with a Keras model

Usage

score_slice_data_dir(slice_data_dir, model_params_dput_file,
  model_h5_weights, score_outdir = NULL, compress_score_rds = FALSE,
  return_score = TRUE, keep_array = FALSE, verbose = FALSE)

Arguments

slice_data_dir	Directory containing sliced image .rds files
model_params_dput_file	Path to file containing dput export of parameters used during training of the model contained in 'model_h5_weights'.
model_h5_weights	Saved h5 weights from a trained keras model, used to score the image tiles.
score_outdir	Path to directory where results of image scoring should be save
compress_score_rds	Should saved score results be compressed
return_score	Should list of score results be returned as an R object by the function
verbose	Should messages about current step being processes be printed to screen?

Value

List of data frame with one row for each tile, containing: List of lists containing model parameters and input 'img_data' data frame with additional column containing model predicted probabilities for each slice

If score_outdir != NULL each list will be saved to score_outdir as an RDS file, compressed if compress_score_rds == TRUE.

treysp/coolit documentation built on June 5, 2019, 5:16 p.m.