Partition into non-overlapping windows with padding if needed. Args: x (tensor): input tokens with [B, H, W, C]. window_size

Window unpartition into original sequences and removing padding. Args: x (tensor): input tokens with [B * num_windows, window_size, w

Window unpartition into original sequences and removing padding. Args: x (tensor): input tokens with [B * num_windows, window_size, w

Args: pic (PIL Image or numpy.ndarray): Image to be converted to tensor. Returns: Tensor: Converted image.

Generate the target heatmap.

Call function to save images.

Generate the target heatmap.

Gets element of the dataset. Args: index: index of the element in the dataset Returns: Single element by index

Args: dim (int): Number of input channels. num_heads (int): Number of attention heads. qkv_bias (bool: I

Args: in_channels (int): Number of input channels. out_channels (int): Number of output channels. bottlen

Args: sampler: Sampler used for subsampling num_replicas (int, optional): Number of processes participating in

Args: dataset_name (str): name of the dataset to be evaluated. distributed (bool): if True, will collect results f

Args: dataset_name: name of the dataset output_dir: output directory to save results for evaluation. eval

Args: dataset_name (str): name of the dataset to be evaluated. distributed (bool): if True, will collect results from

Args: dataset_name: name of the dataset output_dir: output directory to save results for evaluation. eval

Args: dataset_name (str): name of the dataset to be evaluated. distributed (bool): if True, will collect results from

Args: dim (int): Number of input channels. num_heads (int): Number of attention heads. qkv_bias (bool: I

Args: in_channels (int): Number of input channels. out_channels (int): Number of output channels. bottlen

@TODO: Docs. Contribution is welcome.

Returns: int: length of the dataset

Evaluate predictions. Fill self._results with the metrics of the tasks.

load annotation from COCOAPI. Note: bbox:[x1, y1, w, h] Args: img_id: coco image id Returns:

load annotation from COCOAPI. Note: bbox:[x1, y1, w, h] Args: img_id: coco image id Returns:

Load coco person detection results.

Load coco person detection results.

Decay the learning rate with half-cycle cosine after warmup

Args: tensor (Tensor): Tensor image to be normalized. Returns: Tensor: Normalized Tensor image.

Args: img (PIL Image or Tensor): Image to be cropped and resized. Returns: PIL Image or Tensor: Randomly crop

Args: img (PIL Image or Tensor): Image to be flipped. Returns: PIL Image or Tensor: Randomly flipped image.

Args: img (PIL Image or Tensor): Input image. Returns: PIL Image or Tensor: Color jittered image.

Args: img (Tensor): Tensor image to be erased. Returns: img (Tensor): Erased Tensor image.

Calls either forward_train or forward_test depending on whether return_loss=True. Note this setting will change the expected inputs. W

Calls either forward_train or forward_test depending on whether return_loss=True. Note this setting will change the expected inputs. W