mmdet.datasets

datasets

api_wrappers

samplers

transforms

mmdet.engine

hooks

optimizers

runner

schedulers

mmdet.evaluation

functional

mmdet.evaluation.functional.average_precision(recalls,precisions,mode='area')[source]

Calculate average precision (for single or multiple scales).

Parameters

  • recalls (ndarray) – shape (num_scales, num_dets) or (num_dets, )

  • precisions (ndarray) – shape (num_scales, num_dets) or (num_dets, )

  • mode (str) – ‘area’ or ‘11points’, ‘area’ means calculating the areaunder precision-recall curve, ‘11points’ means calculatingthe average precision of recalls at [0, 0.1, …, 1]

Returns

calculated average precision

Return type

float or ndarray

mmdet.evaluation.functional.bbox_overlaps(bboxes1,bboxes2,mode='iou',eps=1e-06,use_legacy_coordinate=False)[source]

Calculate the ious between each bbox of bboxes1 and bboxes2.

Parameters

  • bboxes1 (ndarray) – Shape (n, 4)

  • bboxes2 (ndarray) – Shape (k, 4)

  • mode (str) – IOU (intersection over union) or IOF (intersectionover foreground)

  • use_legacy_coordinate (bool) – Whether to use coordinate system inmmdet v1.x. which means width, height should becalculated as ‘x2 - x1 + 1` and ‘y2 - y1 + 1’ respectively.Note when function is used inVOCDataset, it should beTrue to align with the official implementationhttp://host.robots.ox.ac.uk/pascal/VOC/voc2012/VOCdevkit_18-May-2011.tarDefault: False.

Returns

Shape (n, k)

Return type

ious (ndarray)

mmdet.evaluation.functional.cityscapes_classes() →list[source]

Class names of Cityscapes.

mmdet.evaluation.functional.coco_classes() →list[source]

Class names of COCO.

mmdet.evaluation.functional.coco_panoptic_classes() →list[source]

Class names of COCO panoptic.

mmdet.evaluation.functional.eval_map(det_results,annotations,scale_ranges=None,iou_thr=0.5,ioa_thr=None,dataset=None,logger=None,tpfp_fn=None,nproc=4,use_legacy_coordinate=False,use_group_of=False,eval_mode='area')[source]

Evaluate mAP of a dataset.

Parameters

  • det_results (list[list]) – [[cls1_det, cls2_det, …], …].The outer list indicates images, and the inner list indicatesper-class detected bboxes.

  • annotations (list[dict]) –

    Ground truth annotations where each item ofthe list indicates an image. Keys of annotations are:

    • bboxes: numpy array of shape (n, 4)

    • labels: numpy array of shape (n, )

    • bboxes_ignore (optional): numpy array of shape (k, 4)

    • labels_ignore (optional): numpy array of shape (k, )

  • scale_ranges (list[tuple]| None) – Range of scales to be evaluated,in the format [(min1, max1), (min2, max2), …]. A range of(32, 64) means the area range between (32**2, 64**2).Defaults to None.

  • iou_thr (float) – IoU threshold to be considered as matched.Defaults to 0.5.

  • ioa_thr (float | None) – IoA threshold to be considered as matched,which only used in OpenImages evaluation. Defaults to None.

  • dataset (list[str]| str | None) – Dataset name or dataset classes,there are minor differences in metrics for different datasets, e.g.“voc”, “imagenet_det”, etc. Defaults to None.

  • logger (logging.Logger | str | None) – The way to print the mAPsummary. Seemmengine.logging.print_log() for details.Defaults to None.

  • tpfp_fn (callable | None) – The function used to determine true/false positives. If None,tpfp_default() is used as defaultunless dataset is ‘det’ or ‘vid’ (tpfp_imagenet() in thiscase). If it is given as a function, then this function is usedto evaluate tp & fp. Default None.

  • nproc (int) – Processes used for computing TP and FP.Defaults to 4.

  • use_legacy_coordinate (bool) – Whether to use coordinate system inmmdet v1.x. which means width, height should becalculated as ‘x2 - x1 + 1` and ‘y2 - y1 + 1’ respectively.Defaults to False.

  • use_group_of (bool) – Whether to use group of when calculate TP and FP,which only used in OpenImages evaluation. Defaults to False.

  • eval_mode (str) – ‘area’ or ‘11points’, ‘area’ means calculating thearea under precision-recall curve, ‘11points’ means calculatingthe average precision of recalls at [0, 0.1, …, 1],PASCAL VOC2007 uses11points as default evaluate mode, whileothers are ‘area’. Defaults to ‘area’.

Returns

(mAP, [dict, dict, …])

Return type

tuple

mmdet.evaluation.functional.eval_recalls(gts,proposals,proposal_nums=None,iou_thrs=0.5,logger=None,use_legacy_coordinate=False)[source]

Calculate recalls.

Parameters

  • gts (list[ndarray]) – a list of arrays of shape (n, 4)

  • proposals (list[ndarray]) – a list of arrays of shape (k, 4) or (k, 5)

  • proposal_nums (int | Sequence[int]) – Top N proposals to be evaluated.

  • iou_thrs (float | Sequence[float]) – IoU thresholds. Default: 0.5.

  • logger (logging.Logger | str | None) – The way to print the recallsummary. Seemmengine.logging.print_log() for details.Default: None.

  • use_legacy_coordinate (bool) – Whether use coordinate systemin mmdet v1.x. “1” was added to both height and widthwhich means w, h should becomputed as ‘x2 - x1 + 1` and ‘y2 - y1 + 1’. Default: False.

Returns

recalls of different ious and proposal nums

Return type

ndarray

mmdet.evaluation.functional.evaluateImgLists(prediction_list:list,groundtruth_list:list,args:object,backend_args:Optional[dict]=None,dump_matches:bool=False) →dict[source]

A wrapper of obj:``cityscapesscripts.evaluation.

evalInstanceLevelSemanticLabeling.evaluateImgLists``. Support loadinggroundtruth image from file backend.:param prediction_list: A list of prediction txt file.:type prediction_list: list:param groundtruth_list: A list of groundtruth image file.:type groundtruth_list: list:param args: A global object setting in

obj:cityscapesscripts.evaluation.evalInstanceLevelSemanticLabeling

Parameters

  • backend_args (dict,optional) – Arguments to instantiate thepreifx of uri corresponding backend. Defaults to None.

  • dump_matches (bool) – whether dump matches.json. Defaults to False.

Returns

The computed metric.

Return type

dict

mmdet.evaluation.functional.get_classes(dataset) →list[source]

Get class names of a dataset.

mmdet.evaluation.functional.imagenet_det_classes() →list[source]

Class names of ImageNet Det.

mmdet.evaluation.functional.imagenet_vid_classes() →list[source]

Class names of ImageNet VID.

mmdet.evaluation.functional.objects365v1_classes() →list[source]

Class names of Objects365 V1.

mmdet.evaluation.functional.objects365v2_classes() →list[source]

Class names of Objects365 V2.

mmdet.evaluation.functional.oid_challenge_classes() →list[source]

Class names of Open Images Challenge.

mmdet.evaluation.functional.oid_v6_classes() →list[source]

Class names of Open Images V6.

mmdet.evaluation.functional.plot_iou_recall(recalls,iou_thrs)[source]

Plot IoU-Recalls curve.

Parameters

  • recalls (ndarray orlist) – shape (k,)

  • iou_thrs (ndarray orlist) – same shape asrecalls

mmdet.evaluation.functional.plot_num_recall(recalls,proposal_nums)[source]

Plot Proposal_num-Recalls curve.

Parameters

  • recalls (ndarray orlist) – shape (k,)

  • proposal_nums (ndarray orlist) – same shape asrecalls

mmdet.evaluation.functional.pq_compute_multi_core(matched_annotations_list,gt_folder,pred_folder,categories,backend_args=None,nproc=32)[source]

Evaluate the metrics of Panoptic Segmentation with multithreading.

Same as the function with the same name inpanopticapi.

Parameters

  • matched_annotations_list (list) – The matched annotation list. Eachelement is a tuple of annotations of the same image with theformat (gt_anns, pred_anns).

  • gt_folder (str) – The path of the ground truth images.

  • pred_folder (str) – The path of the prediction images.

  • categories (str) – The categories of the dataset.

  • backend_args (object) – The file client of the dataset. If None,the backend will be set tolocal.

  • nproc (int) – Number of processes for panoptic quality computing.Defaults to 32. Whennproc exceeds the number of cpu cores,the number of cpu cores is used.

mmdet.evaluation.functional.pq_compute_single_core(proc_id,annotation_set,gt_folder,pred_folder,categories,backend_args=None,print_log=False)[source]

The single core function to evaluate the metric of PanopticSegmentation.

Same as the function with the same name inpanopticapi. Only the functionto load the images is changed to use the file client.

Parameters

  • proc_id (int) – The id of the mini process.

  • gt_folder (str) – The path of the ground truth images.

  • pred_folder (str) – The path of the prediction images.

  • categories (str) – The categories of the dataset.

  • backend_args (object) – The Backend of the dataset. If None,the backend will be set tolocal.

  • print_log (bool) – Whether to print the log. Defaults to False.

mmdet.evaluation.functional.print_map_summary(mean_ap,results,dataset=None,scale_ranges=None,logger=None)[source]

Print mAP and results of each class.

A table will be printed to show the gts/dets/recall/AP of each class andthe mAP.

Parameters

  • mean_ap (float) – Calculated fromeval_map().

  • results (list[dict]) – Calculated fromeval_map().

  • dataset (list[str]| str | None) – Dataset name or dataset classes.

  • scale_ranges (list[tuple]| None) – Range of scales to be evaluated.

  • logger (logging.Logger | str | None) – The way to print the mAPsummary. Seemmengine.logging.print_log() for details.Defaults to None.

mmdet.evaluation.functional.print_recall_summary(recalls,proposal_nums,iou_thrs,row_idxs=None,col_idxs=None,logger=None)[source]

Print recalls in a table.

Parameters

  • recalls (ndarray) – calculated frombbox_recalls

  • proposal_nums (ndarray orlist) – top N proposals

  • iou_thrs (ndarray orlist) – iou thresholds

  • row_idxs (ndarray) – which rows(proposal nums) to print

  • col_idxs (ndarray) – which cols(iou thresholds) to print

  • logger (logging.Logger | str | None) – The way to print the recallsummary. Seemmengine.logging.print_log() for details.Default: None.

mmdet.evaluation.functional.voc_classes() →list[source]

Class names of PASCAL VOC.

metrics

mmdet.models

backbones

data_preprocessors

dense_heads

detectors

layers

losses

necks

roi_heads

seg_heads

task_modules

test_time_augs

utils

mmdet.structures

structures

classmmdet.structures.DetDataSample(*,metainfo:Optional[dict]=None,**kwargs)[source]

A data structure interface of MMDetection. They are used as interfacesbetween different components.

The attributes inDetDataSample are divided into several parts:

  • ``proposals``(InstanceData): Region proposals used in two-stage

    detectors.

  • ``gt_instances``(InstanceData): Ground truth of instance annotations.

  • ``pred_instances``(InstanceData): Instances of detection predictions.

  • ``pred_track_instances``(InstanceData): Instances of tracking

    predictions.

  • ``ignored_instances``(InstanceData): Instances to be ignored during

    training/testing.

  • ``gt_panoptic_seg``(PixelData): Ground truth of panoptic

    segmentation.

  • ``pred_panoptic_seg``(PixelData): Prediction of panoptic

    segmentation.

  • ``gt_sem_seg``(PixelData): Ground truth of semantic segmentation.

  • ``pred_sem_seg``(PixelData): Prediction of semantic segmentation.

Examples

>>>importtorch>>>importnumpyasnp>>>frommmengine.structuresimportInstanceData>>>frommmdet.structuresimportDetDataSample
>>>data_sample=DetDataSample()>>>img_meta=dict(img_shape=(800,1196),...pad_shape=(800,1216))>>>gt_instances=InstanceData(metainfo=img_meta)>>>gt_instances.bboxes=torch.rand((5,4))>>>gt_instances.labels=torch.rand((5,))>>>data_sample.gt_instances=gt_instances>>>assert'img_shape'indata_sample.gt_instances.metainfo_keys()>>>len(data_sample.gt_instances)5>>>print(data_sample)

<DetDataSample(

META INFORMATION

DATA FIELDSgt_instances: <InstanceData(

META INFORMATIONpad_shape: (800, 1216)img_shape: (800, 1196)

DATA FIELDSlabels: tensor([0.8533, 0.1550, 0.5433, 0.7294, 0.5098])bboxes:tensor([[9.7725e-01, 5.8417e-01, 1.7269e-01, 6.5694e-01],

[1.7894e-01, 5.1780e-01, 7.0590e-01, 4.8589e-01],[7.0392e-01, 6.6770e-01, 1.7520e-01, 1.4267e-01],[2.2411e-01, 5.1962e-01, 9.6953e-01, 6.6994e-01],[4.1338e-01, 2.1165e-01, 2.7239e-04, 6.8477e-01]])

) at 0x7f21fb1b9190>

) at 0x7f21fb1b9880>
>>>pred_instances=InstanceData(metainfo=img_meta)>>>pred_instances.bboxes=torch.rand((5,4))>>>pred_instances.scores=torch.rand((5,))>>>data_sample=DetDataSample(pred_instances=pred_instances)>>>assert'pred_instances'indata_sample
>>>pred_track_instances=InstanceData(metainfo=img_meta)>>>pred_track_instances.bboxes=torch.rand((5,4))>>>pred_track_instances.scores=torch.rand((5,))>>>data_sample=DetDataSample(...pred_track_instances=pred_track_instances)>>>assert'pred_track_instances'indata_sample
>>>data_sample=DetDataSample()>>>gt_instances_data=dict(...bboxes=torch.rand(2,4),...labels=torch.rand(2),...masks=np.random.rand(2,2,2))>>>gt_instances=InstanceData(**gt_instances_data)>>>data_sample.gt_instances=gt_instances>>>assert'gt_instances'indata_sample>>>assert'masks'indata_sample.gt_instances
>>>data_sample=DetDataSample()>>>gt_panoptic_seg_data=dict(panoptic_seg=torch.rand(2,4))>>>gt_panoptic_seg=PixelData(**gt_panoptic_seg_data)>>>data_sample.gt_panoptic_seg=gt_panoptic_seg>>>print(data_sample)

<DetDataSample(

META INFORMATION

DATA FIELDS_gt_panoptic_seg: <BaseDataElement(

META INFORMATION

DATA FIELDSpanoptic_seg: tensor([[0.7586, 0.1262, 0.2892, 0.9341],

[0.3200, 0.7448, 0.1052, 0.5371]])

) at 0x7f66c2bb7730>

gt_panoptic_seg: <BaseDataElement(

META INFORMATION

DATA FIELDSpanoptic_seg: tensor([[0.7586, 0.1262, 0.2892, 0.9341],

[0.3200, 0.7448, 0.1052, 0.5371]])

) at 0x7f66c2bb7730>

) at 0x7f66c2bb7280>>>> data_sample = DetDataSample()>>> gt_segm_seg_data = dict(segm_seg=torch.rand(2, 2, 2))>>> gt_segm_seg = PixelData(**gt_segm_seg_data)>>> data_sample.gt_segm_seg = gt_segm_seg>>> assert ‘gt_segm_seg’ in data_sample>>> assert ‘segm_seg’ in data_sample.gt_segm_seg

classmmdet.structures.ReIDDataSample(*,metainfo:Optional[dict]=None,**kwargs)[source]

A data structure interface of ReID task.

It’s used as interfaces between different components.

Meta field:
img_shape (Tuple): The shape of the corresponding input image.

Used for visualization.

ori_shape (Tuple): The original shape of the corresponding image.

Used for visualization.

num_classes (int): The number of all categories.

Used for label format conversion.

Data field:

gt_label (LabelData): The ground truth label.pred_label (LabelData): The predicted label.scores (torch.Tensor): The outputs of model.

set_gt_label(value:Union[numpy.ndarray,torch.Tensor,Sequence[numbers.Number],numbers.Number]) →mmdet.structures.reid_data_sample.ReIDDataSample[source]

Set label ofgt_label.

set_gt_score(value:torch.Tensor) →mmdet.structures.reid_data_sample.ReIDDataSample[source]

Set score ofgt_label.

classmmdet.structures.TrackDataSample(*,metainfo:Optional[dict]=None,**kwargs)[source]

A data structure interface of tracking task in MMDetection. It is usedas interfaces between different components.

This data structure can be viewd as a wrapper of multiple DetDataSample tosome extent. Specifically, it only contains a property:video_data_samples which is a list of DetDataSample, each of whichcorresponds to a single frame. If you want to get the property of a singleframe, you must first get the correspondingDetDataSample by indexingand then get the property of the frame, such asgt_instances,pred_instances and so on. As for metainfo, it differs fromDetDataSample in that each value corresponds to the metainfo key is alist where each element corresponds to information of a single frame.

Examples

>>>importtorch>>>frommmengine.structuresimportInstanceData>>>frommmdet.structuresimportDetDataSample,TrackDataSample>>>track_data_sample=TrackDataSample()>>># set the 1st frame>>>frame1_data_sample=DetDataSample(metainfo=dict(...img_shape=(100,100),frame_id=0))>>>frame1_gt_instances=InstanceData()>>>frame1_gt_instances.bbox=torch.zeros([2,4])>>>frame1_data_sample.gt_instances=frame1_gt_instances>>># set the 2nd frame>>>frame2_data_sample=DetDataSample(metainfo=dict(...img_shape=(100,100),frame_id=1))>>>frame2_gt_instances=InstanceData()>>>frame2_gt_instances.bbox=torch.ones([3,4])>>>frame2_data_sample.gt_instances=frame2_gt_instances>>>track_data_sample.video_data_samples=[frame1_data_sample,...frame2_data_sample]>>># set metainfo for track_data_sample>>>track_data_sample.set_metainfo(dict(key_frames_inds=[0]))>>>track_data_sample.set_metainfo(dict(ref_frames_inds=[1]))>>>print(track_data_sample)<TrackDataSample(

META INFORMATIONkey_frames_inds: [0]ref_frames_inds: [1]

DATA FIELDSvideo_data_samples: [<DetDataSample(

META INFORMATIONimg_shape: (100, 100)

DATA FIELDSgt_instances: <InstanceData(

META INFORMATION

DATA FIELDSbbox: tensor([[0., 0., 0., 0.],

[0., 0., 0., 0.]])

) at 0x7f639320dcd0>

) at 0x7f64bd223340>, <DetDataSample(

META INFORMATIONimg_shape: (100, 100)

DATA FIELDSgt_instances: <InstanceData(

META INFORMATION

DATA FIELDSbbox: tensor([[1., 1., 1., 1.],

[1., 1., 1., 1.],[1., 1., 1., 1.]])

) at 0x7f64bd128b20>

) at 0x7f64bd1346d0>]

) at 0x7f64bd2237f0>>>> print(len(track_data_sample))2>>> key_data_sample = track_data_sample.get_key_frames()>>> print(key_data_sample[0].frame_id)0>>> ref_data_sample = track_data_sample.get_ref_frames()>>> print(ref_data_sample[0].frame_id)1>>> frame1_data_sample = track_data_sample[0]>>> print(frame1_data_sample.gt_instances.bbox)tensor([[0., 0., 0., 0.],

[0., 0., 0., 0.]])

>>># Tensor-like methods>>>cuda_track_data_sample=track_data_sample.to('cuda')>>>cuda_track_data_sample=track_data_sample.cuda()>>>cpu_track_data_sample=track_data_sample.cpu()>>>cpu_track_data_sample=track_data_sample.to('cpu')>>>fp16_instances=cuda_track_data_sample.to(...device=None,dtype=torch.float16,non_blocking=False,...copy=False,memory_format=torch.preserve_format)
clone() →mmengine.structures.base_data_element.BaseDataElement[source]

Deep copy the current data element.

Returns

The copy of current data element.

Return type

BaseDataElement

cpu() →mmengine.structures.base_data_element.BaseDataElement[source]

Convert all tensors to CPU in data.

cuda() →mmengine.structures.base_data_element.BaseDataElement[source]

Convert all tensors to GPU in data.

detach() →mmengine.structures.base_data_element.BaseDataElement[source]

Detach all tensors in data.

npu() →mmengine.structures.base_data_element.BaseDataElement[source]

Convert all tensors to NPU in data.

numpy() →mmengine.structures.base_data_element.BaseDataElement[source]

Convert all tensors to np.ndarray in data.

to(*args,**kwargs) →mmengine.structures.base_data_element.BaseDataElement[source]

Apply same name function to all tensors in data_fields.

to_tensor() →mmengine.structures.base_data_element.BaseDataElement[source]

Convert all np.ndarray to tensor in data.

bbox

mask

mmdet.testing

mmdet.visualization

mmdet.utils

classmmdet.utils.AvoidOOM(to_cpu=True,test=False)[source]

Try to convert inputs to FP16 and CPU if got a PyTorch’s CUDA Out ofMemory error. It will do the following steps:

  1. First retry after callingtorch.cuda.empty_cache().

  2. If that still fails, it will then retry by converting inputs

to FP16.

  1. If that still fails trying to convert inputs to CPUs.

In this case, it expects the function to dispatch toCPU implementation.

Parameters

  • to_cpu (bool) – Whether to convert outputs to CPU if get an OOMerror. This will slow down the code significantly.Defaults to True.

  • test (bool) – Skip_ignore_torch_cuda_oom operate that can uselightweight data in unit test, only used intest unit. Defaults to False.

Examples

>>>frommmdet.utils.memoryimportAvoidOOM>>>AvoidCUDAOOM=AvoidOOM()>>>output=AvoidOOM.retry_if_cuda_oom(>>>some_torch_function)(input1,input2)>>># To use as a decorator>>># from mmdet.utils import AvoidCUDAOOM>>>@AvoidCUDAOOM.retry_if_cuda_oom>>>deffunction(*args,**kwargs):>>>returnNone

```

Note

  1. The output may be on CPU even if inputs are on GPU. Processing

    on CPU will slow down the code significantly.

  2. When converting inputs to CPU, it will only look at each argument

    and check if it has.device and.to for conversion. Nestedstructures of tensors are not supported.

  3. Since the function might be called more than once, it has to be

    stateless.

retry_if_cuda_oom(func)[source]

Makes a function retry itself after encountering pytorch’s CUDA OOMerror.

The implementation logic is referred tohttps://github.com/facebookresearch/detectron2/blob/main/detectron2/utils/memory.py

Parameters

func – a stateless callable that takes tensor-like objectsas arguments.

Returns

a callable which retriesfunc if OOM is encountered.

Return type

func

mmdet.utils.all_reduce_dict(py_dict,op='sum',group=None,to_float=True)[source]

Apply all reduce function for python dict object.

The code is modified fromhttps://github.com/Megvii-BaseDetection/YOLOX/blob/main/yolox/utils/allreduce_norm.py.

NOTE: make sure that py_dict in different ranks has the same keys andthe values should be in the same shape. Currently only supportsnccl backend.

Parameters

  • py_dict (dict) – Dict to be applied all reduce op.

  • op (str) – Operator, could be ‘sum’ or ‘mean’. Default: ‘sum’

  • group (torch.distributed.group, optional) – Distributed group,Default: None.

  • to_float (bool) – Whether to convert all values of dict to float.Default: True.

Returns

reduced python dict object.

Return type

OrderedDict

mmdet.utils.allreduce_grads(params,coalesce=True,bucket_size_mb=-1)[source]

Allreduce gradients.

Parameters

  • params (list[torch.Parameters]) – List of parameters of a model

  • coalesce (bool,optional) – Whether allreduce parameters as a whole.Defaults to True.

  • bucket_size_mb (int,optional) – Size of bucket, the unit is MB.Defaults to -1.

mmdet.utils.collect_env()[source]

Collect the information of the running environments.

mmdet.utils.compat_cfg(cfg)[source]

This function would modify some filed to keep the compatibility ofconfig.

For example, it will move some args which will be deprecated to the correctfields.

mmdet.utils.find_latest_checkpoint(path,suffix='pth')[source]

Find the latest checkpoint from the working directory.

Parameters

  • path (str) – The path to find checkpoints.

  • suffix (str) – File extension.Defaults to pth.

Returns

File path of the latest checkpoint.

Return type

latest_path(str | None)

References

1

https://github.com/microsoft/SoftTeacher/blob/main/ssod/utils/patch.py

mmdet.utils.get_caller_name()[source]

Get name of caller method.

mmdet.utils.get_test_pipeline_cfg(cfg:Union[str,mmengine.config.config.ConfigDict]) →mmengine.config.config.ConfigDict[source]

Get the test dataset pipeline from entire config.

Parameters

cfg (str orConfigDict) – the entire config. Can be a configfile or aConfigDict.

Returns

the config of test dataset.

Return type

ConfigDict

mmdet.utils.imshow_mot_errors(*args,backend:str='cv2',**kwargs)[source]

Show the wrong tracks on the input image.

Parameters

backend (str,optional) – Backend of visualization.Defaults to ‘cv2’.

mmdet.utils.log_img_scale(img_scale,shape_order='hw',skip_square=False)[source]

Log image size.

Parameters

  • img_scale (tuple) – Image size to be logged.

  • shape_order (str,optional) – The order of image shape.‘hw’ for (height, width) and ‘wh’ for (width, height).Defaults to ‘hw’.

  • skip_square (bool,optional) – Whether to skip logging for squareimg_scale. Defaults to False.

Returns

Whether to have done logging.

Return type

bool

mmdet.utils.reduce_mean(tensor)[source]

“Obtain the mean of tensor on different GPUs.

mmdet.utils.register_all_modules(init_default_scope:bool=True) →None[source]

Register all modules in mmdet into the registries.

Parameters

init_default_scope (bool) – Whether initialize the mmdet default scope.Wheninit_default_scope=True, the global default scope will beset tommdet, and all registries will build modules from mmdet’sregistry node. To understand more about the registry, please refertohttps://github.com/open-mmlab/mmengine/blob/main/docs/en/tutorials/registry.mdDefaults to True.

mmdet.utils.replace_cfg_vals(ori_cfg)[source]

Replace the string “${key}” with the corresponding value.

Replace the “${key}” with the value of ori_cfg.key in the config. Andsupport replacing the chained ${key}. Such as, replace “${key0.key1}”with the value of cfg.key0.key1. Code is modified from`vars.py< https://github.com/microsoft/SoftTeacher/blob/main/ssod/utils/vars.py>`_ # noqa: E501

Parameters

ori_cfg (mmengine.config.Config) – The origin config with “${key}” generated from a file.

Returns

The config with “${key}” replaced by the corresponding value.

Return type

updated_cfg [mmengine.config.Config]

mmdet.utils.setup_cache_size_limit_of_dynamo()[source]

Setup cache size limit of dynamo.

Note: Due to the dynamic shape of the loss calculation andpost-processing parts in the object detection algorithm, thesefunctions must be compiled every time they are run.Setting a large value for torch._dynamo.config.cache_size_limitmay result in repeated compilation, which can slow down trainingand testing speed. Therefore, we need to set the default value ofcache_size_limit smaller. An empirical value is 4.

mmdet.utils.setup_multi_processes(cfg)[source]

Setup multi-processing environment variables.

mmdet.utils.split_batch(img,img_metas,kwargs)[source]

Split data_batch by tags.

Code is modified from<https://github.com/microsoft/SoftTeacher/blob/main/ssod/utils/structure_utils.py> # noqa: E501

Parameters

  • img (Tensor) – of shape (N, C, H, W) encoding input images.Typically these should be mean centered and std scaled.

  • img_metas (list[dict]) – List of image info dict where each dicthas: ‘img_shape’, ‘scale_factor’, ‘flip’, and may also contain‘filename’, ‘ori_shape’, ‘pad_shape’, and ‘img_norm_cfg’.For details on the values of these keys, seemmdet.datasets.pipelines.Collect.

  • kwargs (dict) – Specific to concrete implementation.

Returns

a dict that data_batch splited by tags,

such as ‘sup’, ‘unsup_teacher’, and ‘unsup_student’.

Return type

data_groups (dict)

mmdet.utils.sync_random_seed(seed=None,device='cuda')[source]

Make sure different ranks share the same seed.

All workers must call this function, otherwise it will deadlock.This method is generally used inDistributedSampler,because the seed should be identical across all processesin the distributed group.

In distributed sampling, different ranks should sample non-overlappeddata in the dataset. Therefore, this function is used to make sure thateach rank shuffles the data indices in the same order basedon the same seed. Then different ranks could use different indicesto select non-overlapped data from the same data list.

Parameters

  • seed (int,Optional) – The seed. Default to None.

  • device (str) – The device where the seed will be put on.Default to ‘cuda’.

Returns

Seed to be used.

Return type

int

mmdet.utils.update_data_root(cfg,logger=None)[source]

Update data root according to env MMDET_DATASETS.

If set env MMDET_DATASETS, update cfg.data_root according toMMDET_DATASETS. Otherwise, using cfg.data_root as default.

Parameters

  • cfg (Config) – The model config need to modify

  • logger (logging.Logger | str | None) – the way to print msg