Computes pairwise intersection-over-area between box collections. Intersection-over-area (ioa) between two boxes box1 and box2 is defined as

Loading boxes and labels from AVA bounding boxes csv files. Args: cfg (CfgNode): config. video_name (str): name of the given

Prepares the image for network input with format of CHW RGB float

Main function to spawn the train and test process.

Calculate the median epoch time among the recorded epochs.

Merge data from precomputed and ground-truth boxes dictionaries. Args: pred_dict (dict): a dict which maps from `frame_idx` to a list

This function converts target class indices to one-hot vectors, given the number of classes. Args: targets (loader): Class labels

Non maximum suppression. This op greedily selects a subset of detection bounding boxes, pruning away boxes that have high IOU (intersection o

Non maximum suppression. This op greedily selects a subset of detection bounding boxes, pruning away boxes that have high IOU (intersection o

Convert BN parameters to Sub-BN parameters if model contains Sub-BNs. Args: checkpoint_sd (OrderedDict): source dict of parameters.

Convert BN parameters to Sub-BN parameters if model contains Sub-BNs. Args: checkpoint_sd (OrderedDict): source dict of parameters.

Pad the given image with the size of pad_size. Args: image (array): image to pad. pad_size (int): size to pad. order

Compute the number of parameters. Args: model (model): model to count the number of parameters.

For classification: Perform mutli-view testing that uniformly samples N clips from a video along its temporal axis. For each clip, it tak

Visualize video inputs with wrong predictions on Tensorboard. Args: vis_loader (loader): video visualization loader. model (m

Given all predictions and all true labels, plot histograms of top-k most frequently predicted classes for each true class. Args:

Log schedule.

Extract layer names and numpy-like fancy indexing from a string. Args: layer_ls (list of strs): list of strings containing data about

Add the new task to task queue. Args: task (TaskInfo object): task object that contain the necessary info

Put task into task queue for prediction and visualization. Args: task (TaskInfo object): task object that contain

Convert the video from its original fps to the target_fps. If the video support selective decoding (contain decoding information in the video

Decode the video with PyAV decoder. Args: container (container): PyAV container. start_pts (int): the starting Presentation T

Generates a random square bbox based on lambda value. Args: img_shape (tuple): Image shape as tuple lam (float): Cutmix lamb

Reset the metric.

Run demo visualization. Args: cfg (CfgNode): configs. Details can be found in slowfast/config/defaults.py frame_p

Run model visualization (weights, activations and model inputs) and visualize them on Tensorboard. Args: vis_loader (loader): vid

Perform color saturation on the given image. Args: var (float): variance. image (array): image to perform color saturation.

Perfrom saturation jittering on the input images. The channels of images should be in order BGR. Args: var (float): jitter ratio

Perform color saturation on the list of given images. Args: var (float): variance. images (list): list of images to perform c

Scale box coordinates in x and y dimensions. Args: boxlist: BoxList holding N boxes y_scale: float x_scale: float Returns:

Scale the short side of the box to size. Args: size (int): size to scale the image. boxes (ndarray): bounding boxes to peform

Convert time index (in second) to frame index. 0: 900 30: 901

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 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`

Given the start and end frame index, sample num_samples frames between the start and end with equal interval. Args: frames (tenso

Normalize a given tensor by subtracting the mean and dividing the std. Args: tensor (tensor): tensor to normalize. mean (tens

If video_meta is not empty, perform temporal selective decoding to sample a clip from the video with TorchVision decoder. If video_meta is em

Perform the video training for one epoch. Args: train_loader (loader): video training loader. model (model): the video model

If separate==True, the transforms are returned as a tuple of 3 separate transforms for use in a mixing dataset that passes * all data th

Update predictions and labels. Args: preds (tensor): model output predictions. labels (tensor): labels.

Draw predicted labels on video inputs and visualize all incorrectly classified videos in the current batch. Args:

Transpose the dimension from `channel` x `height` x `width` to `height` x `width` x `channel`. Args: image (array): image to

Transpose the dimension from `height` x `width` x `channel` to `channel` x `height` x `width`. Args: image (array): image to

Construct PyTorchVideo Charades video loader. Load Charades data (frame paths, labels, etc. ) to Charades Dataset object. The dataset cou

Construct the Kinetics video loader with a given csv file. The format of the csv file is: ``` path_to_video_1 label_1 path_to_vid

Construct PyTorchVideo Something-Something v2 SSv2 video loader. Load SSv2 data (frame paths, labels, etc. ) to SSv2 Dataset object. The

How many results are ready to be returned.

Returns the prediction results for the current task. Args: task (TaskInfo object): task object that contain

Return bounding boxes predictions as a tensor. Args: task (TaskInfo object): task object that contain the

Visualize the localization maps on their corresponding inputs as heatmap, using Grad-CAM. Args: inputs (list of t

Args: img (PIL Image): Image to be cropped and resized. Returns: PIL Image: Randomly cropped and resized imag

Given the video index, return the list of frames, label, and video index if the video can be fetched and decoded successfully, otherw

Generate corresponding clips, boxes, labels and metadata for given idx. Args: idx (int): the video index provided by the

Given the video index, return the list of frames, label, and video index if the video frames can be fetched. Args:

Given the video index, return the list of frames, label, and video index if the video frames can be fetched. Args:

Given the video index, return the list of frames, label, and video index if the video frames can be fetched. Args:

overall_iters (int): the overall number of iterations of one epoch. cfg (CfgNode): configs. mode (str): `train`, `val`, or `t

Construct tensors to store the predictions and labels. Expect to get num_clips predictions from each video, and calculate the metrics

Args: window_size (int): size of the max length of the deque.

Args: epoch_iters (int): the overall number of iterations of one epoch. cfg (CfgNode): configs.

Args: max_iter (int): the max number of iteration of the current epoch. cfg (CfgNode): configs.

Constructs box collection. Args: data: a numpy array of shape [N, 4] representing box coordinates Raises: ValueE

Initialized PerImageEvaluation by evaluation parameters. Args: num_groundtruth_classes: Number of ground truth object classes

Constructs box collection. Args: box_data: a numpy array of shape [N, 4] representing box coordinates mask_data: a numpy

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

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

Predict Worker for Detectron2. Args: cfg (CfgNode): configs. Details can be found in slow

Visualization Worker for AsyncVis. Args: video_vis (VideoVisualizer object): object with tools for visualizat

Args: cfg (CfgNode): configs. Details can be found in slowfast/config/defaults.py async_vis (AsyncVis

See https://github.com/facebookresearch/detectron2/blob/master/detectron2/utils/visualizer.py for more details. Args:

Args: num_classes (int): total number of classes. class_names_path (str): path to json file that maps class names to

Args: cfg (CfgNode): configs. Details can be found in slowfast/config/defaults.py

Args: cfg (CfgNode): configs. Details can be found in slowfast/config/defaults.py

Args: cfg (CfgNode): configs. Details can be found in slowfast/config/defaults.py

Args: cfg (CfgNode): configs. Details can be found in slowfast/config/defaults.py gpu_id (Optional[in