[Bug]

[southeastboss]

Prerequisite

• I have searched Issues and Discussions but cannot get the expected help.
• I have read the FAQ documentation but cannot get the expected help.
• The bug has not been fixed in the latest version (master) or latest version (3.x).

Task

I have modified the scripts/configs, or I'm working on my own tasks/models/datasets.

Branch

3.x branch https://github.com/open-mmlab/mmdetection/tree/3.x

Environment

sys.platform: linux
Python: 3.8.10 (default, Jun 4 2021, 15:09:15) [GCC 7.5.0]
CUDA available: True
numpy_random_seed: 2147483648
GPU 0: NVIDIA GeForce RTX 3060
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 11.3, V11.3.109
GCC: gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
PyTorch: 1.11.0+cu113
PyTorch compiling details: PyTorch built with:

• GCC 7.3
• C++ Version: 201402
• Intel(R) Math Kernel Library Version 2020.0.0 Product Build 20191122 for Intel(R) 64 architecture applications
• Intel(R) MKL-DNN v2.5.2 (Git Hash a9302535553c73243c632ad3c4c80beec3d19a1e)
• OpenMP 201511 (a.k.a. OpenMP 4.5)
• LAPACK is enabled (usually provided by MKL)
• NNPACK is enabled
• CPU capability usage: AVX2
• CUDA Runtime 11.3
• NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86
• CuDNN 8.2
• Magma 2.5.2
• Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=11.3, CUDNN_VERSION=8.2.0, CXX_COMPILER=/opt/rh/devtoolset-7/root/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=OFF, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF,

TorchVision: 0.12.0+cu113
OpenCV: 4.6.0
MMEngine: 0.3.2
MMDetection: 3.0.0rc5+92d03df

Reproduces the problem - code sample

base = [
'../base/models/faster-rcnn_r50_fpn.py',
'../base/default_runtime.py',
'../base/datasets/semi_coco_detection.py',
'../base/schedules/schedule_1x.py'
]

detector = base.model
detector.data_preprocessor = dict(
type='DetDataPreprocessor',
mean=[103.530, 116.280, 123.675],
std=[1.0, 1.0, 1.0],
bgr_to_rgb=False,
pad_size_divisor=32)
detector.backbone = dict(
type='ResNet',
depth=50,
num_stages=4,
out_indices=(0, 1, 2, 3),
frozen_stages=1,
norm_cfg=dict(type='BN', requires_grad=False),
norm_eval=True,
style='caffe',
init_cfg=dict(
type='Pretrained',
checkpoint='open-mmlab://detectron2/resnet50_caffe'))

model = dict(
delete=True,
type='SoftTeacher',
detector=detector,
data_preprocessor=dict(
type='MultiBranchDataPreprocessor',
data_preprocessor=detector.data_preprocessor),
semi_train_cfg=dict(
freeze_teacher=True,
sup_weight=1.0,
unsup_weight=4.0,
pseudo_label_initial_score_thr=0.5,
rpn_pseudo_thr=0.9,
cls_pseudo_thr=0.9,
reg_pseudo_thr=0.02,
jitter_times=10,
jitter_scale=0.06,
min_pseudo_bbox_wh=(1e-2, 1e-2)),
semi_test_cfg=dict(predict_on='teacher'))

custom_hooks = [dict(type='MeanTeacherHook')]
val_cfg = dict(type='TeacherStudentValLoop')
work_dir='work_dir/faster_rcnn_semi_detection'

Reproduces the problem - command or script

python tools/train.py configs/faster_rcnn/faster_rcnn_semi_detection.py

Reproduces the problem - error message

Traceback (most recent call last):
File "tools/train.py", line 130, in
main()
File "tools/train.py", line 126, in main
runner.train()
File "/root/miniconda3/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1654, in train
self._train_loop = self.build_train_loop(
File "/root/miniconda3/lib/python3.8/site-packages/mmengine/runner/runner.py", line 1446, in build_train_loop
loop = LOOPS.build(
File "/root/miniconda3/lib/python3.8/site-packages/mmengine/registry/registry.py", line 454, in build
return self.build_func(cfg, *args, **kwargs, registry=self)
File "/root/miniconda3/lib/python3.8/site-packages/mmengine/registry/build_functions.py", line 135, in build_from_cfg
raise type(e)(
ValueError: class EpochBasedTrainLoop in mmengine/runner/loops.py: class ConcatDataset in mmengine/dataset/dataset_wrapper.py: class CocoDataset in mmdet/datasets/coco.py: need at least one array to concatenate

Additional information

I try to use semi-supervised object detection in my Dataset I follow the docs in https://mmdetection.readthedocs.io/en/3.x/user_guides/semi_det.html
I prepare and split my dataset as blow
mmdetection
├── data
│ ├── coco
│ │ ├── annotations
│ │ │ ├── image_info_unlabeled2017.json
│ │ │ ├── instances_train2017.json
│ │ │ ├── instances_unlabeled2017.json
│ │ │ ├── instances_val2017.json
│ │ ├── train2017
│ │ ├── unlabeled2017
│ │ ├── val2017

and using the config show in the code sample part,
but it arise error show in the error message part.

[ZwwWayne]

@Czm369 please to take a look

[ZwwWayne]

Please see if you can solve you issue following the solution in #9610.

[southeastboss]

I solve this problem with change the config. Now I use the config which inherit the config in configs/soft_teacher/soft-teacher_faster-rcnn_r50-caffe_fpn_180k_semi-0.1-coco.py and change the key of 'num_classes', 'labeled_dataset.ann_file', 'unlabeled_dataset.ann_file', and add key of 'metainfo' in each dataset dict part.

the metainfo as below
metainfo=dict(
classes=('a', 'b', 'c'),
palette=[(255, 0, 0), (0, 255, 0), (0, 0, 255)]),

I see the config in configs/soft_teacher/soft-teacher_faster-rcnn_r50-caffe_fpn_180k_semi-0.1-coco.py use the train_cfg as blow
train_cfg = dict(type='IterBasedTrainLoop', max_iters=180000, val_interval=5000)
so in the semi_detection can we use other type train_cfg? I have not see in the user_guides.
so I think the user_guides in https://mmdetection.readthedocs.io/en/3.x/user_guides/semi_det.html is not complete.
I hope the group of mmdetection can improve the docs and we can use the mmdetection more convenient.
thank you!

[Czm369]

Because semi-supervised object detection involves labeled dataset and unlabeled dataset, the sizes of the two data sets are different and it is difficult to define epoch, so only supported IterBasedTrainLoop.

[Sundragon1993]

@Czm369 In case I train a traditional supervise method like RTMDet using ConcatDataset type, could you confirm that the mmdet v3 doesn't support using EpochBasedTrainLoop?