Sets model layers as trainable if their names match the given regular expression.

Run detection on images in the given directory.

Takes RGB images with 0-255 values and subtraces the mean pixel and converts it to float. Expects image colors in RGB order.

Takes attributes of an image and puts them in one 1D array. Use parse_image_meta() to parse the values back. image_id: An int ID of t

Train the model.

Reformats the detections of one image from the format of the neural network output to a format suitable for use in the rest of the application

Reformats the detections of one image from the format of the neural network output to a format suitable for use in the rest of the application

Converts a mask generated by the neural network into a format similar to its original shape. mask: [height, width] of type float. A small, typ

Converts a mask generated by the neural network into a format similar to its original shape. mask: [height, width] of type float. A small, typ

Args: total: total number of steps.

A generator that returns images and corresponding target class ids, bounding box deltas, and masks. dataset_handler: The Data

model_dir: Directory to save training logs and trained weights

This class should not be instantiated.

Returns a value inside the range [-1, 0]. The higher the value, closer are the boxes.

Compute Intersection over Union of ground truth and predicted masks. Args: gt_masks (torch.IntTensor((img_height, img_width, nb_gt_masks)

Applies the given deltas to the given boxes. Args: boxes: [batch_size, N, 4] where each row is y1, x1, y2, x2 deltas: [batch_size

Compute refinement needed to transform box to gt_box. box and gt_box are [N, (y1, x1, y2, x2)]

Parses an image info Numpy array to its components. See to_numpy() for more details.

Returns a shorter version of object names for cleaner display.

Compute loss for single image according to: https://www.kaggle.com/c/data-science-bowl-2018#evaluation

Computes IoU overlaps between two sets of boxes. boxes1, boxes2: [N, (y1, x1, y2, x2)]. For better performance, pass the largest set first an

Displays configurations.

Display ground truth and prediction instances on the same image.

Display the given image and the top few class masks.

Scans all the weights in the model and returns a list of tuples that contain stats about each weight.

Draw 3-pixel width bounding boxes on the given image array. color: list of 3 int values for RGB.

Draw bounding boxes and segmentation masks with differnt customizations. boxes: [N, (y1, x1, y2, x2, class_id)] in image coordinates. ref

anchors: [n, (y1, x1, y2, x2)] list of anchors in image coordinates. proposals: [n, 4] the same anchors but refined to fit objects better.

Resizes mini masks back to image size. Reverses the change of minimize_mask(). See inspect_data.ipynb notebook for more details.

Compute bounding boxes from masks. mask: [height, width, num_instances]. Mask pixels are either 1 or 0. Returns: bbox array [num_instances, (

Returns the image_id given the idx.

Map an internal class ID to the corresponding class ID in the source dataset.

Determines which augmenters to apply to masks.

Return the path of the image.

Return a link to the image in its source Website or details about the image that help looking it up or debugging it. Override for you

Loads configurations from YAML files, then create utilitaire configurations. Freeze config and display it.

Generate instance masks for an image. Returns: masks: A bool array of shape [height, width, instance count] with one mask p

Takes a source class ID and returns the int class ID assigned to it. For example: dataset.map_source_class_id("coco.12") -> 23

Resize masks to a smaller version to cut memory load. Mini-masks can then resized back to image scale using expand_masks() See inspect_data.i

Takes a list of images and modifies them to the format expected as an input to the neural network. images: List of image matricies [height,wid

CUDA proxy

Draw a grid showing how ground truth objects are classified. gt_class_ids: [N] int. Ground truth class IDs pred_class_id: [N] int. Predicted c

Draw the precision-recall curve. AP: Average precision at IoU >= 0.5 precisions: list of precision values recalls: list of recall values