Closed nuonepeaceyy closed 1 year ago
If I replace the './dcff_yolov5_s_cspdarknet.py'
with './dcff_faster_rcnn_resnet50_fpn.py'
in
_base_ = [
'./dcff_yolov5_s_cspdarknet.py',
# 'mmdet::_base_/datasets/coco_detection.py',
# 'mmdet::_base_/schedules/schedule_2x.py',
'mmyolo::_base_/default_runtime.py'
]
this problem no longer exists, I don't understand the reason, they all exist under the same path and I don't know if my config file is correct.
When mmengine(openmmlab) raise error when a config can't be found, it usually print the absolute path of the config. Can you check the absolute path of the config?
When mmengine(openmmlab) raise error when a config can't be found, it usually print the absolute path of the config. Can you check the absolute path of the config?
Thank you for your answer, after changing the base file with mmyolo::yolov5/yolov5_s-v61_syncbn_fast_8xb16-300e_coco.py
, the problem is solved.
Checklist
Describe the question you meet
I try to configure the relevant yolov5s config files according to the example in configs/pruning/mmdet/dcff, but in fact, I don't have a very clear idea about how to do this. When I try to get the config template of target_pruning_ratio by using
python tools/pruning/get_channel_units.py configs/pruning/mmdet/dcff/dcff_yolov5_s_cspdarknet_1b32.py --choice
, it always getFileNotFoundError: [Errno 2] No such file or directory: '/tmp/tmpnbjbrug0/tmpi2cqdj_l.py'
Post related information
pip list | grep "mmcv\|mmrazor\|^torch"
mmcv 2.0.0rc4 mmdet 3.0.0rc5 mmrazor 1.0.0rc2 /home/MMLAB/mmrazor mmyolo 0.4.0 torch 1.12.1 torchaudio 0.12.1 torchvision 0.13.1