Mimics tensorflow's 'SAME' padding.

Stores image metas.

Handles nuclei dataset.

MRCNN detection output. rois: [N, (y1, x1, y2, x2)] Bounding boxes in pixels class_ids: [N] Integer class IDs for each bounding b

Head of the FPN that generates masks.

Builds the model of Region Proposal Network. anchors_per_location: number of anchors per pixel in the feature map anchor_stride: Controls the

Stores execution configuration. Usage: Initialization: Config.load_default(config_fn) Config.merge(other_fn)

This class is used as an empty object where each instance will store different attributes according to the configuration tree.

The base class for dataset classes. To use it, create a new class that adds functions specific to the dataset you want to use. For example:

Encapsulates the Mask RCNN model functionality.

Metaclass used to block modifications to Config attributes if the configuration is frozen.