Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Fix] Add BN in FPN to avoid loss Nan in MVXNet #2282

Merged
merged 5 commits into from
Feb 18, 2023
Merged

[Fix] Add BN in FPN to avoid loss Nan in MVXNet #2282

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
10f22e4
add BN to avoid loss nan
JingweiZhang12 Feb 15, 2023
e7ea664
add comments about cfg
JingweiZhang12 Feb 15, 2023
87aea53
update mvxnet configs and readme
JingweiZhang12 Feb 17, 2023
19c9a37
update checkpoint and log
JingweiZhang12 Feb 17, 2023
8f3e1c2
update link in metafile
JingweiZhang12 Feb 17, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 10 additions & 3 deletions configs/mvxnet/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,9 +20,16 @@ We implement MVX-Net and provide its results and models on KITTI dataset.

### KITTI

| Backbone | Class | Lr schd | Mem (GB) | Inf time (fps) | mAP | Download |
| :-------------------------------------------------------------------: | :-----: | :--------: | :------: | :------------: | :---: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| [SECFPN](./dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class.py) | 3 Class | cosine 80e | 6.7 | | 63.22 | [model](https://download.openmmlab.com/mmdetection3d/v1.0.0_models/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class_20210831_060805-83442923.pth) \| [log](https://download.openmmlab.com/mmdetection3d/v1.0.0_models/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class_20210831_060805.log.json) |
| Backbone | Class | Lr schd | Mem (GB) | Inf time (fps) | mAP11/mAP40 | Download |
| :-------------------------------------------------------------------: | :-----: | :--------: | :------: | :------------: | :---------: | :----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
| [SECFPN](./dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class.py) | 3 Class | cosine 40e | 7.4 | | 63.3/63.0 | [model](https://download.openmmlab.com/mmdetection3d/v1.0.0_models/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class/dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class-5a3a671a.pth) \| [log](https://download.openmmlab.com/mmdetection3d/v1.0.0_models/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class/20230215_133955.log) |

Detailed performance on KITTI 3D detection is as follows, evaluated by AP11 and AP40 metric:

| | Easy | Moderate | Hard |
| --------- | :-----: | :------: | :-----: |
| 3d (AP11) | 72.6428 | 63.3217 | 59.6384 |
| 3d (AP40) | 73.8799 | 63.0147 | 59.1020 |

## Citation

Expand Down
2 changes: 2 additions & 0 deletions ...d_secfpn_adamw_2x8_80e_kitti-3d-3class.py → ...d_secfpn_adamw_2x8_40e_kitti-3d-3class.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,8 @@
type='FPN',
in_channels=[256, 512, 1024, 2048],
out_channels=256,
# make the image features more stable numerically to avoid loss nan
norm_cfg=dict(type='BN', requires_grad=False),
num_outs=5),
pts_voxel_layer=dict(
max_num_points=-1,
Expand Down
10 changes: 5 additions & 5 deletions configs/mvxnet/metafile.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,14 +17,14 @@ Collections:
Version: v0.5.0

Models:
- Name: dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class
- Name: dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class
In Collection: MVX-Net
Config: configs/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class.py
Config: configs/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class.py
Metadata:
Training Memory (GB): 6.7
Training Memory (GB): 7.4
Results:
- Task: 3D Object Detection
Dataset: KITTI
Metrics:
mAP: 63.22
Weights: https://download.openmmlab.com/mmdetection3d/v1.0.0_models/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class/dv_mvx-fpn_second_secfpn_adamw_2x8_80e_kitti-3d-3class_20210831_060805-83442923.pth
mAP: 63.0
Weights: https://download.openmmlab.com/mmdetection3d/v1.0.0_models/mvxnet/dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class/dv_mvx-fpn_second_secfpn_adamw_2x8_40e_kitti-3d-3class-5a3a671a.pth