Convenience function for accesssing box coordinates. Returns: a numpy array of shape [N, 4] representing box corners

Accesses data associated with the specified field in the box collection. Args: field: a string parameter used to speficy a related field to

Return number of boxes held in collections.

Reset the individual meters.

Add data to a specified field. Args: field: a string parameter used to speficy a related field to be accessed. field_data: a numpy ar

Convenience function for accessing masks. Returns: a numpy array of shape [N, height, width] representing masks

Gather boxes from BoxList according to indices and return new BoxList. By default, gather returns boxes corresponding to the input index list, as

Constructor. Args: categories: A list of dicts, each of which has the following keys - 'id': (required) an integer id uniquely iden

Get path for storing checkpoints. Args: path_to_job (string): the path to the folder of the current job.

Return all non-box fields.

Load the checkpoint from the given file. If inflation is True, inflate the 2D Conv weights from the checkpoint to 3D Conv. Args:

ResBlock class constructs redisual blocks. More details can be found in: Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun.

Given scale, ratio, height and width, return sampled coordinates of the videos.

Converts a BoxList containing 'masks' into a BoxMaskList. Args: boxlist: An np_box_list.BoxList object. Returns: An np_box_mask_list.Box

Gather boxes from np_box_mask_list.BoxMaskList according to indices. By default, gather returns boxes corresponding to the input index list, as w

Convert the image to gray scale. Args: image (tensor): image to convert to gray scale. Dimension is `channel` x `height`

Start to record time.

Scale the short side of the image to size. Args: size (int): size to scale the image. image (array): image to perform short s

The `__init__` method of any subclass should also contain these arguments. Args: dim_in (int): the channel dimension of

The `__init__` method of any subclass should also contain these arguments. ResNetRoIHead takes p pathways as input where

Adds groundtruth for a single image to be used for evaluation. Args: image_id: A unique string/integer identifier for the image. grou

Computes area of boxes. Args: boxes: Numpy array with shape [N, 4] holding N boxes Returns: a numpy array with shape [N*1] representing

Computes area of masks. Args: masks: Numpy array with shape [N, height, width] holding N masks. Masks values are of type np.uint8 and val

Blend two images with a given weight alpha. Args: images1 (tensor): the first images to be blended, the dimension is `num

Calculate and log the final PSI-AVA metrics.

Calculate the current median value of the deque.

Get the grayscale for the input images. The channels of images should be in order BGR. Args: images (tensor): the input images fo

Copies the extra fields of boxlist_to_copy_from to boxlist_to_copy_to. Args: boxlist_to_copy_to: BoxList to which extra fields are copied.

Adds detections for a single image to be used for evaluation. Args: image_id: A unique string/integer identifier for the image. detec

Add a new scalar value to each of the task's deques.

Calculate the average epoch time among the recorded epochs.

Perform center crop on input images. Args: size (int): size of the cropped height and width. image (array): the image to perf

Clears the state to prepare for a fresh evaluation.

Peform crop on the bounding boxes given the offsets. Args: boxes (ndarray or None): bounding boxes to peform crop. The dimension

Stochastic Depth per sample.

Evaluate the model on the val set. Args: val_loader (loader): data loader to provide validation data. model (model): model to

Evaluates detections and returns a dictionary of metrics.

Filter to keep only boxes and masks with score exceeding a given threshold. This op keeps the collection of boxes and masks whose corresponding sco

Filter to keep only boxes with score exceeding a given threshold. This op keeps the collection of boxes whose corresponding scores are greater th

Get the last checkpoint from the checkpointing folder. Args: path_to_job (string): the path to the folder of the current job.

Given the configs, retrieve the specified lr policy function. Args: lr_policy (string): the learning rate policy to use for the job.

Compute statistics for the current model given the config. Args: model (model): model to perform analysis. cfg (CfgNode): con

Determines if the given directory contains a checkpoint. Args: path_to_job (string): the path to the folder of the current job.

Compute pairwise intersection areas between boxes. Args: boxes1: a numpy array with shape [N, 4] holding N boxes boxes2: a numpy array with

Compute pairwise intersection areas between masks. Args: masks1: a numpy array with shape [N, height, width] holding N masks. Masks value

Stop to record time.

Get the time for the last epoch.

Loads label map proto. Args: path: path to StringIntLabelMap proto text file. Returns: a StringIntLabelMapProto

Log the stats of the current epoch. Args: cur_epoch (int): the number of current epoch.

Log the stats. Args: cur_epoch (int): the current epoch. cur_iter (int): the current iteration.

Returns a unique identifier for a video id & timestamp.

Save json for the specific task. Args: cur_epoch (int): the number of current epoch.

Sort boxes and associated fields according to a scalar field. A common use case is reordering the boxes according to descending scores. Args:

Convert the Sub-BN paprameters to normal BN parameters in a state dict. There are two copies of BN layers in a Sub-BN implementation: `bn.bn`

Convert the model into training mode while keep layers freezed.

Update the current stats. Args: preds (tensor): prediction embedding. keep_box(tensor): tensor of boolean to

Args: x: (B, D, H, W, C) window_size (tuple[int]): window size Returns: windows: (B*num_windows, window_size*window_

Args: num_splits (int): number of splits. args (list): other arguments.

Args: dim_in (int): the channel dimensions of the input. ratio (float): the channel reduction ratio for squeeze.

Labels true/false positives of detections of an image across all classes. Args: detected_boxes: A float numpy array of shape [N, 4], repres

Labels boxes detected with the same class from the same image as tp/fp. Args: detected_boxes: A numpy array of shape [N, 4] representing de

Builds a SlowFast model. The first pathway is the Slow pathway and the second pathway is the Fast pathway. Args:

Calculate the aggregated mean and stds. Args: means (tensor): mean values. stds (tensor): standard deviations

Return a process group based on gloo backend, containing all the ranks The result is cached. Returns: (group): pytorch dist group

Returns numpy arrays belonging to class with index `class_index`. Args: detected_boxes: A numpy array containing detected boxes. dete

Return a dummy input for model analysis with batch size 1. The input is used for analyzing the model (counting flops and activations etc.

Computes overlaps and scores between detected and groudntruth boxes. Args: detected_boxes: A numpy array of shape [N, 4] representing detec

This function performs preprocessing for the input images and corresponding boxes for one clip. Args: imgs (tens

This function performs preprocessing for the input images and corresponding boxes for one clip with opencv as backend. Args:

Check whether data fullfills the format of N*[ymin, xmin, ymax, xmin]. Args: data: a numpy array of shape [N, 4] representing box coordinat

Load frame paths and annotations from files Args: cfg (CfgNode): config

Padding all the tensors from different GPUs to the largest ones. Args: tensor (tensor): tensor to pad. group (group): pytorch

Removes entries with invalid boxes. A box is invalid if either its xmax is smaller than its xmin, or its ymax is smaller than its ymin.