microglia_analyzer.ma_worker module

class microglia_analyzer.ma_worker.MicrogliaAnalyzer(logging_f=None)

Bases: object

analyze_graph()

as_tsv(identifier)

bindings_from_yolo(yolo_bboxes_str)

bindings_to_yolo()

classify_microglia()

get_classification_version()

get_mask(show_garbage)

get_segmentation_version()

reset_classification()

reset_segmentation()

segment_microglia()

set_calibration(pixel_size, unit)

Setter of the calibration. Before editing the internal state, checks that the pixel size is a float and Pint is used to check the unit.

set_classification_model(path, use='best', reload=False)

Checks that the path corresponds to a folder. This folder must contain a “confusion_matrix.png” file to verify that the training is complete. In there, there must be a “weights” folder, containing either ‘best.pt’ or ‘last.pt’.

Parameters:

• path (-) – Path of the model’s folder (containing ‘results.csv’ and ‘weights’).

• use (-) – Either ‘best’ or ‘last’, to use either ‘best.pt’ or ‘last.pt’.

• reload (-) – Whether to force the reload of the model from the online repo.

set_input_image(image)

Setter of the input image. Checks that the image is 2D before using it.

set_min_surface(min_size)

set_proba_threshold(threshold)

set_segmentation_model(path, use='best')

Checks that the path is a folder. In the folder, we search for the file “best.keras” or “last.keras”. To verify that the training was complete, we also check for the presence of “training_history.png”.

set_working_directory(path)

Checks that the directory exists before setting it. Also outputs a warning in the logs if it is not empty.