Builder of box coder.

Args: img (Tensor): Input images of shape (N, C, H, W). Typically these should be mean centered and std scaled.

Check if is one of the norms.

Load checkpoint from a file or URI. Args: model (Module): Module to load checkpoint. filename (str): Accept local filepath, URL, `

Extract a multi-scale single image tensor from a multi-scale batch tensor based on batch index. Note: The default value of detach is True, be

Update data root according to env MMDET_DATASETS. If set env MMDET_DATASETS, update cfg.data_root according to MMDET_DATASETS. Otherwise, usi

Build backbone.

Evaluate mAP of a dataset. Args: det_results (list[list]): [[cls1_det, cls2_det, ...], ...]. The outer list indicates images,

Calculate recalls. Args: gts (list[ndarray]): a list of arrays of shape (n, 4) proposals (list[ndarray]): a list of arrays of sha

Filter results using score threshold and topk candidates. Args: scores (Tensor): The scores, shape (num_bboxes, K). score_thr (fl

Return the class containing the corresponding attributes according to the fpn_test_step_names.

Draw bboxes and class labels (with scores) on an image. Args: img (str | ndarray): The image to be displayed. bboxes (ndarray): B

Inference image(s) with the detector. Args: model (nn.Module): The loaded detector. imgs (str/ndarray or list[str/ndarray] or tup

Log image size. Args: img_scale (tuple): Image size to be logged. shape_order (str, optional): The order of image shape.

Test function without test time augmentation. Args: imgs (list[torch.Tensor]): List of multiple images img_metas (lis

Generate sparse anchors according to the ``prior_idxs``. Args: prior_idxs (Tensor): The index of corresponding anchors

Transpose and gather feature according to index. Args: feat (Tensor): Target feature map. ind (Tensor): Target coord index.

Replace ResNet50 with ResNet18 in config.

Builder of IoU calculator.

Build neck.

Deprecated sample_per_gpu in cfg.data.

Apply transformation `pred_bboxes` to `boxes`. Args: bboxes (torch.Tensor): Basic boxes. Shape (B, N, 4) or (N, 4) pr

flip tensor base on flip_direction. Args: src_tensor (Tensor): input feature map, shape (B, C, H, W). flip_direction (str): The f

Format the results to json (standard format for COCO evaluation). Args: results (list[tuple | numpy.ndarray]): Testing results of

Convert Mask to ndarray.. Args: mask (:obj:`BitmapMasks` or :obj:`PolygonMasks` or torch.Tensor or np.ndarray): The mask to be co

Merge augmented mask prediction. Args: aug_masks (list[ndarray]): shape (n, #class, h, w) img_shapes (list[ndarray]): shape (3, )

Regress the bbox for the predicted class. Used in Cascade R-CNN. Args: rois (Tensor): Rois from `rpn_head` or last stage

Set user-defined new property.

Simple forward test function.

bbox post-processing method. The boxes would be rescaled to the original image scale and do the nms operation. Usually with_nms is Fa

Handle empty batch dimension to adaptive_avg_pool2d. Args: input (tensor): 4D tensor. output_size (int, tuple[int,int]): the targ

Calculate overlap between two set of bboxes. FP16 Contributed by https://github.com/open-mmlab/mmdetection/pull/4889 Note: Assume bbo

Build PyTorch DataLoader. In distributed training, each GPU/process has a dataloader. In non-distributed training, there is only one dataload

Builder of IoU calculator.

Check if norm layer is in correct train state.

Apply transformation `pred_bboxes` to `boxes`. Args: bboxes (torch.Tensor): Basic boxes.Shape (B, N, 4) or (N, 4) pre

Directly extract features from the backbone+neck.

Extract features from multiple images. Args: imgs (list[torch.Tensor]): A list of images. The images are augmente

Find the latest checkpoint from the working directory. Args: path(str): The path to find checkpoints. suffix(str): File extension

Get custom activation of cls_score. Args: cls_score (torch.Tensor): The prediction with shape (N, C + 2). Returns:

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

make plugins for block. Args: in_channels (int): Input channels of plugin. plugins (list[dict]): List of plugins cfg

Returns: list: Contains all keys in meta_info_fields.

nn.Module: the normalization layer named "norm2"

Parse the package dependencies listed in a requirements file but strips specific versioning information. Args: fname (str): path to r

Refine bboxes during training. Args: rois (Tensor): Shape (n*bs, 5), where n is image number per GPU, and bs is t

see :func:`BaseInstanceMasks.resize`

Setup multi-processing environment variables.

See :func:`BaseInstanceMasks.to_tensor`.

Args: img_size (int): Input image size. patch_size (int): Patch size. in_chans (int): Number of input ima

Box head forward function used in both training and testing.

Box head forward function used in both training and testing.

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

Paste instance masks according to boxes. This implementation is modified from https://github.com/facebookresearch/detectron2/ Args:

Create sample results that can be passed to BBoxHead.get_targets.

Mask head forward function used in both training and testing.

Create a dummy onnx::NonMaxSuppression op while exporting to ONNX. This function helps exporting to onnx with batch and multiclass NMS op. It

See :func:`BaseInstanceMasks.crop`.

See :func:`BaseInstanceMasks.crop_and_resize`.

Encode bitmap mask to RLE code. Args: mask_results (list | tuple[list]): bitmap mask results. In mask scoring rcnn, mask_resu

Directly extract features from the backbone+neck.

Generate the coordinate. Args: featmap_sizes (tuple): The feature to be calculated, of shape (N, C, W, H). device (st

Get category ids of concatenated dataset by index. Args: idx (int): Index of data. Returns: list[int]: All c

Get COCO category ids by index. Args: idx (int): Index of data. Returns: list[int]: All categories in the im

Get user-defined property.

Get the valid radios of feature maps of all level.

Interpolate the `source` to the shape of the `target`. The `source` must be a Tensor, but the `target` can be a Tensor or a np.ndarray with t

In distributed setting, this function only download checkpoint at local rank 0.

Make divisible function. This function rounds the channel number to the nearest value that can be divisible by the divisor. It is taken from

Merge augmented detection bboxes and scores. Args: aug_bboxes (list[Tensor]): shape (n, 4*#class) aug_scores (list[Tensor] or Non

nn.Module: normalization layer after the third convolution layer

Prepare results dict for pipeline.

Makes a function retry itself after encountering pytorch's CUDA OOM error. The implementation logic is referred to https://gi

Update a dict to `data_fields`. Args: data (dict): A dict contains annotations of image or model predictions. Def

Convert objects of various python types to :obj:`torch.Tensor`. Supported types are: :class:`numpy.ndarray`, :class:`torch.Tensor`, :class:`S

Box head forward function used in both training and testing.

Check data in fields from two results are same.

A context which ignores CUDA OOM exception from pytorch. Code is modified from <https://github.com/facebookresearch/detectron2/blob/main/dete

Load checkpoint from somewhere (modelzoo, file, url). Args: filename (str): Accept local filepath, URL, ``torchvision://xxx``,