Return one normalized image tensor with box and keypoint targets.

Return fake series by column name.

Creates the matcher. Args: cost_class: Relative weight of the classification error in the matching cost. cost_bbox: R

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

Initializes the model. Parameters: backbone: torch module of the backbone to be used. See backbone.py transformer: to

Multi-Scale Deformable Attention Module :param d_model hidden dimension :param n_levels number of feature levels :param n_heads

ch_in, ch_out, shortcut, groups, kernels, expand.

Args: in_channels: Channels in each input feature map level. out_channels: Number of channels in the output feature m

Create the initializer. Args: dim: Query conditioning dimensionality. num_queries: Number of query tokens to instanti

Initialise stub with optional class names, label list, and keypoint flag.

Store initialization kwargs for optional inspection.

Initialize the fake response with content and metadata.

Store values for ``notna().any()`` checks.

Store column data for ``_build_metric_groups``.