Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
sguokaiwei@163.com 6246a11263 | 1 year ago | |
---|---|---|
.. | ||
configs | 1 year ago | |
mmcv_custom | 1 year ago | |
mmdet/models/backbones | 1 year ago | |
README.md | 1 year ago |
We add ConvNeXt model and config files to Swin Detection.
Our code has been tested with commit 6a979e2
. Please refer to README.md for installation and dataset preparation instructions.
name | Pretrained Model | Method | Lr Schd | box mAP | mask mAP | #params | FLOPs | Fine-tuned Model |
---|---|---|---|---|---|---|---|---|
ConvNeXt-T | ImageNet-1K | Mask R-CNN | 3x | 46.2 | 41.7 | 48M | 262G | model |
ConvNeXt-T | ImageNet-1K | Cascade Mask R-CNN | 3x | 50.4 | 43.7 | 86M | 741G | model |
ConvNeXt-S | ImageNet-1K | Cascade Mask R-CNN | 3x | 51.9 | 45.0 | 108M | 827G | model |
ConvNeXt-B | ImageNet-1K | Cascade Mask R-CNN | 3x | 52.7 | 45.6 | 146M | 964G | model |
ConvNeXt-B | ImageNet-22K | Cascade Mask R-CNN | 3x | 54.0 | 46.9 | 146M | 964G | model |
ConvNeXt-L | ImageNet-22K | Cascade Mask R-CNN | 3x | 54.8 | 47.6 | 255M | 1354G | model |
ConvNeXt-XL | ImageNet-22K | Cascade Mask R-CNN | 3x | 55.2 | 47.7 | 407M | 1898G | model |
To train a detector with pre-trained models, run:
# single-gpu training
python tools/train.py <CONFIG_FILE> --cfg-options model.pretrained=<PRETRAIN_MODEL> [other optional arguments]
# multi-gpu training
tools/dist_train.sh <CONFIG_FILE> <GPU_NUM> --cfg-options model.pretrained=<PRETRAIN_MODEL> [other optional arguments]
For example, to train a Cascade Mask R-CNN model with a ConvNeXt-T
backbone and 8 gpus, run:
tools/dist_train.sh configs/convnext/cascade_mask_rcnn_convnext_tiny_patch4_window7_mstrain_480-800_giou_4conv1f_adamw_3x_coco_in1k.py 8 --cfg-options model.pretrained=https://dl.fbaipublicfiles.com/convnext/convnext_tiny_1k_224.pth
More config files can be found at configs/convnext
.
# single-gpu testing
python tools/test.py <CONFIG_FILE> <DET_CHECKPOINT_FILE> --eval bbox segm
# multi-gpu testing
tools/dist_test.sh <CONFIG_FILE> <DET_CHECKPOINT_FILE> <GPU_NUM> --eval bbox segm
This code is built using mmdetection, timm libraries, and BeiT, Swin Transformer repositories.
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》