return random split of the dataset into training_set, validation_set, and test_set. Args: train (int): percentage of dataset to b

Performs Non-Maximum Suppression (NMS) on inference results Returns: detections with shape: nx6 (x1, y1, x2, y2, conf, cls)

Return intersection-over-union (Jaccard index) of boxes. Both sets of boxes are expected to be in (x1, y1, x2, y2) format. Arguments:

Compute the average precision, given the recall and precision curves. Source: https://github.com/rbgirshick/py-faster-rcnn. # Arguments

Manage all aspects of inference for both networks (yolo and light). Args: img (numpy.ndarray): image to be processed. save_path (

Manage all aspects of inference for both networks (yolo and light). Args: img (numpy.ndarray): image to be processed. save_path (

returns an image's matrix without alpha channel based on its index in the dict_dataset keys. Args: idx (int): index for image in

returns an image's absolute path based on its index in the dict_dataset keys. Args: idx (int): index for image in the dict_datase

Creates kmeans-evolved anchors from training dataset Arguments: path: path to dataset *.yaml, or a loaded dataset n:

Manages bbox to mouse emulator input conversion and clicking. Args: bbox (list): list of four integers representing the pixel coordinates

Decorator to make all processes in distributed training wait for each local_master to do something.

Receives a session folder and updates de dataset's dictionary with one keypair for each image; Note that the function does not return anything. I

generates dictionary Args: root_path (str): root directory that contains the session folders. classes (list): list of