Apply state to `model` from this object

extract the appropriate t index for a batch of indices

If `resume` is True, this method attempts to resume from the last checkpoint, if exists. Otherwise, load checkpoint from the given pa

Add config for DiffusionInst

Returns: torch.nn.Module: It now calls :func:`detectron2.modeling.build_model`. Overwrite it if you'd like a dif

Return the ema state stored in `model`

Arguments: box_cls (Tensor): tensor of shape (batch_size, num_proposals, K). The tensor predicts the classificati

:param features :param weights: [w0, w1, ...] :param bias: [b0, b1, ...] :return:

Visualizes predictions on frames of the input video. Args: video (cv2.VideoCapture): a :class:`VideoCapture` object, who

Save model state from `model` to this object

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

Create the criterion. Parameters: num_classes: number of object categories, omitting the special no-object category m

Args: cfg (CfgNode): num_gpus (int): if 0, will run on CPU

Return an activation function given a string

Apply transformation `deltas` (dx, dy, dw, dh) to `boxes`. Args: deltas (Tensor): transformation deltas of shape (N, k*4

Apply ema stored in `model` to model and returns a function to restore the weights are applied

Create evaluator(s) for a given dataset. This uses the special metadata "evaluator_type" associated with each builtin dataset.

Build a list of default hooks, including timing, evaluation, checkpointing, lr scheduling, precise BN, writing events. Retur

Create a list of :class:`TransformGen` from config. Returns: list[TransformGen]

Check that the datasets have consistent metadata. Args: key (str): a metadata key dataset_names (list[str]): a list of datase

cosine schedule as proposed in https://openreview.net/forum?id=-NEXDKk8gZ

Resize the output instances. The input images are often resized when entering an object detector. As a result, we often need the outputs

boxes: (N, 4) shape

Generalized IoU from https://giou.stanford.edu/ The boxes should be in [x0, y0, x1, y1] format Returns a [N, M] pairwise matrix, where

Args: gt_classes: a long tensor of shape R that contains the gt class label of each proposal. num_fed_loss_classes: m

Get frequency weight for each class sorted by class id. We now calcualte freqency weight using image_count to the power freq_weight_power.

Initialize the weights in backbone. Args: pretrained (str, optional): Path to pre-trained weights. Defaults to Non

:param features :param weights: [w0, w1, ...] :param bias: [b0, b1, ...] :return:

:param gt_boxes: (cx, cy, w, h), normalized :param num_proposals:

Normalize, pad and batch the input images.

Args: image (np.ndarray): an image of shape (H, W, C) (in BGR order). This is the format used by OpenCV.

Create configs and perform basic setups.

This function disables printing when not in master process