Calculate Mean Intersection and Union (mIoU) Args: results (list[ndarray]): List of prediction segmentation maps. gt_seg_maps (li

nn.Module: normalization layer after the first convolution layer

Prepare results dict for pipeline.

Randomly sample an img_scale when ``ratio_range`` is specified. A ratio will be randomly sampled from the range specified by ``ratio_

Randomly sample an img_scale when ``ratio_range`` is specified. A ratio will be randomly sampled from the range specified by ``ratio_

Draw `result` over `img`. Args: img (str or Tensor): The image to be displayed. result (Tensor): The semantic segment

Convert the model into training mode whill keeping the normalization layer freezed.

Run forward function and calculate loss for auxiliary head in training.

Check branches configuration.

Build :obj:`ConcatDataset by.

Convert trainId to id for cityscapes.

Convert segmentation label to onehot. Args: seg_label (Tensor): Segmentation label of shape (N, H, W). num_classes (i

Run forward function and calculate loss for decode head in inference.

Run forward function and calculate loss for decode head in training.

Create a superset of inputs needed to run test or train batches. Args: input_shape (tuple): input batch dimensions nu

Create a superset of inputs needed to run test or train batches. Args: input_shape (tuple): input batch dimensions n

Create a superset of inputs needed to run test or train batches. Args: input_shape (tuple): input batch dimensions n

Evaluation in Cityscapes protocol. Args: results (list): Testing results of the dataset. logger (logging.Logger | str

Expand onehot labels to match the size of prediction.

Find the predefined segmentor config directory.

Load a configuration as a python module.

Grab configs necessary to create a segmentor. These are deep copied to allow for safe modification of parameters without influencing other te

Initialize ``auxiliary_head``

Initialize ``decode_head``

Check and initialize input transforms. The in_channels, in_index and input_transform must match. Specifically, when input_transform i

Build multiple branch.

Build fuse layer.

Make each layer.

Build one branch.

Make stem layer for ResNet.

Pad images according to ``self.size``.

Pad masks according to ``results['pad_shape']``.

Parse the raw outputs (losses) of the network. Args: losses (dict): Raw output of the network, which usually contain

Randomly sample an img_scale according to ``ratio_range`` and ``multiscale_mode``. If ``ratio_range`` is specified, a ratio will be s

Randomly sample an img_scale according to ``ratio_range`` and ``multiscale_mode``. If ``ratio_range`` is specified, a ratio will be s

Resize images with ``results['scale']``.

Resize images with ``results['scale']``.

Resize semantic segmentation map with ``results['scale']``.

Resize semantic segmentation map with ``results['scale']``.

Placeholder for augmentation test.

Brightness distortion.

Build pixel sampler for segmentation map.

CAM feature classification.

Collect results with CPU.

Collect results with GPU.

DropBlock. See https://arxiv.org/pdf/1810.12890.pdf DropBlock with an experimental gaussian noise option. This layer has been tested on a few tra

DropBlock. See https://arxiv.org/pdf/1810.12890.pdf DropBlock with an experimental gaussian noise option. Simplied from above without concern for

Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks). This is the same as the DropConnect impl I created for Ef

Placeholder for extract features from images.

Place holder to format result to dataset specific output.

Format the results into dir (standard format for Cityscapes evaluation). Args: results (list): Testing results of the dat

Forward function.

Forward function.

Forward function for testing. Args: inputs (list[Tensor]): List of multi-level img features. prev_output (Tensor): Th

Forward function for testing, ignore se_loss.

Forward function for testing, only ``pam_cam`` is used.

Get annotation by index. Args: idx (int): Index of data. Returns: dict: Annotation info of specified index.

Get class names of current dataset. Args: classes (Sequence[str] | str | None): If classes is None, use default C

Randomly get a crop bounding box.

Sample points for testing. Find ``num_points`` most uncertain points from ``uncertainty_map``. Args: seg_logits (Tensor)

Sample points for training. Sample points in [0, 1] x [0, 1] coordinate space based on their uncertainty. The uncertainties are calcu

Hue distortion.

Initialize the weights in backbone and heads. Args: pretrained (str, optional): Path to pre-trained weights. Defa

Initialize the weights in backbone. Args: pretrained (str, optional): Path to pre-trained weights. Defaults to No

Initialize weight of later layer.

Calculate intersection and Union. Args: pred_label (ndarray): Prediction segmentation map. label (ndarray): Ground truth segmenta

Load annotation from directory. Args: img_dir (str): Path to image directory img_suffix (str): Suffix of images.