Forward function for `SinePositionalEncoding`. Args: mask (Tensor): ByteTensor mask. Non-zero values representing

Forward function for `LearnedPositionalEncoding`. Args: mask (Tensor): ByteTensor mask. Non-zero values representing

Args: x (Tensor): Has shape (B, H*W, C_in). input_size (tuple[int]): The spatial shape of x, arrange as (H, W).

Forward function for `TransformerCoder`. Returns: Tensor: forwarded results with shape [num_query, bs, embed_dims].

Forward function for `TransformerDecoder`. Args: query (Tensor): Input query with shape `(num_query, bs, embed_di

Forward function for `Transformer`. Args: x (Tensor): Input query with shape [bs, c, h, w] where c = embed_dims.

Forward function for `TransformerDecoder`. Args: query (Tensor): Input query with shape `(num_query, bs, embed_di

Forward function for `Transformer`. Args: mlvl_feats (list(Tensor)): Input queries from different level. Each ele

Forward function for `DynamicConv`. Args: param_feature (Tensor): The feature can be used to generate the paramet

Forward function. Args: feats (list[Tensor]): Multi scale Features from the upstream network, each is a 4D-tensor

Forward function. Args: feats (list[Tensor]): Features from the upstream network, each is a 4D-tensor.

Args: feats (list[Tensor]): Feature maps of each level. Each has shape of (batch_size, c, h, w). Returns

Args: feats (list[Tensor]): Feature maps of each level. Each has shape of [bs, c, h, w]. img_metas (l

Args: feats (list[Tensor]): Feature maps of each level. Each has shape of [bs, c, h, w]. img_metas (l

Forward function. Args: pred (torch.Tensor): The prediction, has a shape (n, *). target (torch.Tensor): The label of

Forward function.

Forward function. Args: pred (torch.Tensor): The prediction. target (torch.Tensor): The learning label of the predict

Dummy forward function.

Dummy forward function.

Test segment without test-time aumengtation. Only the output of last decoder layers was used. Args: inputs (list[Tensor]

Forward function for training mode. Args: x (list[Tensor]): Multi-level features from the upstream network, each

Forward function for training mode. Args: x (list[Tensor]): Multi-level features from the upstream network, each

Generate grid points of multiple feature levels. Args: featmap_sizes (list[tuple]): List of feature map sizes in

Returns a dictionary of info about the object.

Returns a dictionary of info about the object.

Initialize the transformer weights.

Initialize weights.

Initialize weights.

Initialize weights.

Loss Name. This function must be implemented and will return the name of this loss function. This name will be used to combine differ

Loss function for outputs from a single decoder layer. Args: cls_scores (Tensor): Mask score logits from a single decoder layer

Loss function for outputs from a single decoder layer. Args: cls_scores (Tensor): Mask score logits from a single decoder layer

Obtain tight bounding boxes of binary masks. Args: masks (Tensor): Binary mask of shape (n, h, w). Returns: Tensor: Bboxe wi

Calculate the CrossEntropy loss for masks. Args: pred (torch.Tensor): The prediction with shape (N, C), C is the number of cl

torch.Tensor: concatenated positive and negative boxes