Run an active learning loop for iterative segmentation refinement

Orchestrates a human-in-the-loop (HITL) active learning workflow: segment, review, correct, retrain, repeat.

Usage

sg_active_learning_loop(
  image,
  model = "cellpose",
  n_rounds = 5L,
  patches_per_round = 10L,
  patch_size = 256L,
  initial_model = "cyto3"
)

Arguments

image: An sg_image object.
model: Character. Segmentation backend to use. Currently only "cellpose" is supported. Default "cellpose".
n_rounds: Integer. Number of active learning rounds. Default 5L.
patches_per_round: Integer. Patches sampled per round. Default 10L.
patch_size: Integer. Patch size in pixels. Default 256L.
initial_model: Character. Initial model name. Default "cyto3".

Value

An sg_hitl_result S3 object with elements $rounds (list of per-round results), $final_model (trained model path or object), and $metrics (training metrics across rounds).

Examples

# \donttest{
# Requires Python + Cellpose and interactive session:
# result <- sg_active_learning_loop(image, n_rounds = 3L)
# }