Comments (3)
Hi, I never tested the result on nuscenes mini data. Could you reproduce the results on the nuscenes val set?
from uvtr.
Hi, I managed to free up enough space to download the whole dataset and I got similar results.
user@gpu-cluster:~/src/mmdet$ python3 extra_tools/test.py configs/uvtr/multi_modality/uvtr_m_v0075_r101_h5.py ckpts/uvtr/uvtr_m_v0075_r101_h5.pth --eval=bbox
/usr/local/lib/python3.8/dist-packages/mmcv/__init__.py:20: UserWarning: On January 1, 2023, MMCV will release v2.0.0, in which it will remove components related to the training process and add a data transformation module. In addition, it will rename the package names mmcv to mmcv-lite and mmcv-full to mmcv. See https://github.com/open-mmlab/mmcv/blob/master/docs/en/compatibility.md for more details.
warnings.warn(
/usr/local/lib/python3.8/dist-packages/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3190.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
load checkpoint from local path: ckpts/uvtr/uvtr_m_v0075_r101_h5.pth
2023-04-30 12:26:43,335 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.0.conv2 is upgraded to version 2.
2023-04-30 12:26:43,337 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.1.conv2 is upgraded to version 2.
2023-04-30 12:26:43,338 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.2.conv2 is upgraded to version 2.
2023-04-30 12:26:43,339 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.3.conv2 is upgraded to version 2.
2023-04-30 12:26:43,340 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.4.conv2 is upgraded to version 2.
2023-04-30 12:26:43,341 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.5.conv2 is upgraded to version 2.
2023-04-30 12:26:43,342 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.6.conv2 is upgraded to version 2.
2023-04-30 12:26:43,343 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.7.conv2 is upgraded to version 2.
2023-04-30 12:26:43,344 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.8.conv2 is upgraded to version 2.
2023-04-30 12:26:43,345 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.9.conv2 is upgraded to version 2.
2023-04-30 12:26:43,346 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.10.conv2 is upgraded to version 2.
2023-04-30 12:26:43,347 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.11.conv2 is upgraded to version 2.
2023-04-30 12:26:43,349 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.12.conv2 is upgraded to version 2.
2023-04-30 12:26:43,350 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.13.conv2 is upgraded to version 2.
2023-04-30 12:26:43,351 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.14.conv2 is upgraded to version 2.
2023-04-30 12:26:43,352 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.15.conv2 is upgraded to version 2.
2023-04-30 12:26:43,353 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.16.conv2 is upgraded to version 2.
2023-04-30 12:26:43,354 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.17.conv2 is upgraded to version 2.
2023-04-30 12:26:43,355 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.18.conv2 is upgraded to version 2.
2023-04-30 12:26:43,356 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.19.conv2 is upgraded to version 2.
2023-04-30 12:26:43,357 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.20.conv2 is upgraded to version 2.
2023-04-30 12:26:43,358 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.21.conv2 is upgraded to version 2.
2023-04-30 12:26:43,359 - root - INFO - ModulatedDeformConvPack img_backbone.layer3.22.conv2 is upgraded to version 2.
2023-04-30 12:26:43,360 - root - INFO - ModulatedDeformConvPack img_backbone.layer4.0.conv2 is upgraded to version 2.
2023-04-30 12:26:43,362 - root - INFO - ModulatedDeformConvPack img_backbone.layer4.1.conv2 is upgraded to version 2.
2023-04-30 12:26:43,364 - root - INFO - ModulatedDeformConvPack img_backbone.layer4.2.conv2 is upgraded to version 2.
[ ] 0/6019, elapsed: 0s, ETA:/usr/local/lib/python3.8/dist-packages/torch/nn/functional.py:4227: UserWarning: Default grid_sample and affine_grid behavior has changed to align_corners=False since 1.3.0. Please specify align_corners=True if the old behavior is desired. See the documentation of grid_sample for details.
warnings.warn(
[ ] 1/6019, 0.3 task/s, elapsed: 3s, ETA: 18116s ~/src/mmdet/mmdet3d/core/bbox/coders/nms_free_coder.py:67: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
self.post_center_range = torch.tensor(
[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 6019/6019, 2.1 task/s, elapsed: 2800s, ETA: 0s
Formating bboxes of pts_bbox
Start to convert detection format...
[>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>] 6019/6019, 40.5 task/s, elapsed: 149s, ETA: 0s
Results writes to /tmp/tmp2zhts1f2/results/pts_bbox/results_nusc.json
Evaluating bboxes of pts_bbox
mAP: 0.1348
mATE: 0.8543
mASE: 0.3209
mAOE: 0.6170
mAVE: 0.8633
mAAE: 0.2407
NDS: 0.2778
Eval time: 83.0s
Per-class results:
Object Class AP ATE ASE AOE AVE AAE
car 0.142 0.875 0.180 0.249 0.971 0.183
truck 0.049 0.992 0.265 0.324 0.876 0.229
bus 0.017 1.112 0.248 0.555 2.468 0.604
trailer 0.021 1.100 0.283 0.908 0.330 0.135
construction_vehicle 0.002 1.001 0.529 1.144 0.130 0.228
pedestrian 0.295 0.541 0.323 0.683 0.649 0.316
motorcycle 0.177 0.681 0.296 0.629 1.075 0.194
bicycle 0.253 0.663 0.311 0.703 0.408 0.038
traffic_cone 0.225 0.773 0.394 nan nan nan
barrier 0.166 0.803 0.380 0.359 nan nan
{'pts_bbox_NuScenes/car_AP_dist_0.5': 0.0019, 'pts_bbox_NuScenes/car_AP_dist_1.0': 0.0591, 'pts_bbox_NuScenes/car_AP_dist_2.0': 0.191, 'pts_bbox_NuScenes/car_AP_dist_4.0': 0.316, 'pts_bbox_NuScenes/car_trans_err': 0.8745, 'pts_bbox_NuScenes/car_scale_err': 0.1798, 'pts_bbox_NuScenes/car_orient_err': 0.2486, 'pts_bbox_NuScenes/car_vel_err': 0.9711, 'pts_bbox_NuScenes/car_attr_err': 0.1828, 'pts_bbox_NuScenes/mATE': 0.8543, 'pts_bbox_NuScenes/mASE': 0.3209, 'pts_bbox_NuScenes/mAOE': 0.617, 'pts_bbox_NuScenes/mAVE': 0.8633, 'pts_bbox_NuScenes/mAAE': 0.2407, 'pts_bbox_NuScenes/truck_AP_dist_0.5': 0.0, 'pts_bbox_NuScenes/truck_AP_dist_1.0': 0.0057, 'pts_bbox_NuScenes/truck_AP_dist_2.0': 0.0585, 'pts_bbox_NuScenes/truck_AP_dist_4.0': 0.1328, 'pts_bbox_NuScenes/truck_trans_err': 0.9924, 'pts_bbox_NuScenes/truck_scale_err': 0.2653, 'pts_bbox_NuScenes/truck_orient_err': 0.3236, 'pts_bbox_NuScenes/truck_vel_err': 0.8758, 'pts_bbox_NuScenes/truck_attr_err': 0.2292, 'pts_bbox_NuScenes/construction_vehicle_AP_dist_0.5': 0.0, 'pts_bbox_NuScenes/construction_vehicle_AP_dist_1.0': 0.0, 'pts_bbox_NuScenes/construction_vehicle_AP_dist_2.0': 0.0002, 'pts_bbox_NuScenes/construction_vehicle_AP_dist_4.0': 0.0097, 'pts_bbox_NuScenes/construction_vehicle_trans_err': 1.0013, 'pts_bbox_NuScenes/construction_vehicle_scale_err': 0.5295, 'pts_bbox_NuScenes/construction_vehicle_orient_err': 1.144, 'pts_bbox_NuScenes/construction_vehicle_vel_err': 0.1305, 'pts_bbox_NuScenes/construction_vehicle_attr_err': 0.2276, 'pts_bbox_NuScenes/bus_AP_dist_0.5': 0.0, 'pts_bbox_NuScenes/bus_AP_dist_1.0': 0.0, 'pts_bbox_NuScenes/bus_AP_dist_2.0': 0.0076, 'pts_bbox_NuScenes/bus_AP_dist_4.0': 0.0606, 'pts_bbox_NuScenes/bus_trans_err': 1.1117, 'pts_bbox_NuScenes/bus_scale_err': 0.2478, 'pts_bbox_NuScenes/bus_orient_err': 0.5546, 'pts_bbox_NuScenes/bus_vel_err': 2.4679, 'pts_bbox_NuScenes/bus_attr_err': 0.6041, 'pts_bbox_NuScenes/trailer_AP_dist_0.5': 0.0, 'pts_bbox_NuScenes/trailer_AP_dist_1.0': 0.0, 'pts_bbox_NuScenes/trailer_AP_dist_2.0': 0.0116, 'pts_bbox_NuScenes/trailer_AP_dist_4.0': 0.072, 'pts_bbox_NuScenes/trailer_trans_err': 1.1001, 'pts_bbox_NuScenes/trailer_scale_err': 0.283, 'pts_bbox_NuScenes/trailer_orient_err': 0.908, 'pts_bbox_NuScenes/trailer_vel_err': 0.3297, 'pts_bbox_NuScenes/trailer_attr_err': 0.1348, 'pts_bbox_NuScenes/barrier_AP_dist_0.5': 0.011, 'pts_bbox_NuScenes/barrier_AP_dist_1.0': 0.0915, 'pts_bbox_NuScenes/barrier_AP_dist_2.0': 0.2187, 'pts_bbox_NuScenes/barrier_AP_dist_4.0': 0.3434, 'pts_bbox_NuScenes/barrier_trans_err': 0.8034, 'pts_bbox_NuScenes/barrier_scale_err': 0.3798, 'pts_bbox_NuScenes/barrier_orient_err': 0.3595, 'pts_bbox_NuScenes/barrier_vel_err': nan, 'pts_bbox_NuScenes/barrier_attr_err': nan, 'pts_bbox_NuScenes/motorcycle_AP_dist_0.5': 0.0253, 'pts_bbox_NuScenes/motorcycle_AP_dist_1.0': 0.133, 'pts_bbox_NuScenes/motorcycle_AP_dist_2.0': 0.2365, 'pts_bbox_NuScenes/motorcycle_AP_dist_4.0': 0.3129, 'pts_bbox_NuScenes/motorcycle_trans_err': 0.6813, 'pts_bbox_NuScenes/motorcycle_scale_err': 0.2961, 'pts_bbox_NuScenes/motorcycle_orient_err': 0.629, 'pts_bbox_NuScenes/motorcycle_vel_err': 1.0745, 'pts_bbox_NuScenes/motorcycle_attr_err': 0.1937, 'pts_bbox_NuScenes/bicycle_AP_dist_0.5': 0.0458, 'pts_bbox_NuScenes/bicycle_AP_dist_1.0': 0.1917, 'pts_bbox_NuScenes/bicycle_AP_dist_2.0': 0.3496, 'pts_bbox_NuScenes/bicycle_AP_dist_4.0': 0.4252, 'pts_bbox_NuScenes/bicycle_trans_err': 0.6635, 'pts_bbox_NuScenes/bicycle_scale_err': 0.311, 'pts_bbox_NuScenes/bicycle_orient_err': 0.7029, 'pts_bbox_NuScenes/bicycle_vel_err': 0.4081, 'pts_bbox_NuScenes/bicycle_attr_err': 0.0378, 'pts_bbox_NuScenes/pedestrian_AP_dist_0.5': 0.1057, 'pts_bbox_NuScenes/pedestrian_AP_dist_1.0': 0.264, 'pts_bbox_NuScenes/pedestrian_AP_dist_2.0': 0.3596, 'pts_bbox_NuScenes/pedestrian_AP_dist_4.0': 0.4512, 'pts_bbox_NuScenes/pedestrian_trans_err': 0.5412, 'pts_bbox_NuScenes/pedestrian_scale_err': 0.3233, 'pts_bbox_NuScenes/pedestrian_orient_err': 0.6827, 'pts_bbox_NuScenes/pedestrian_vel_err': 0.649, 'pts_bbox_NuScenes/pedestrian_attr_err': 0.3156, 'pts_bbox_NuScenes/traffic_cone_AP_dist_0.5': 0.0235, 'pts_bbox_NuScenes/traffic_cone_AP_dist_1.0': 0.1444, 'pts_bbox_NuScenes/traffic_cone_AP_dist_2.0': 0.2996, 'pts_bbox_NuScenes/traffic_cone_AP_dist_4.0': 0.4316, 'pts_bbox_NuScenes/traffic_cone_trans_err': 0.7734, 'pts_bbox_NuScenes/traffic_cone_scale_err': 0.3937, 'pts_bbox_NuScenes/traffic_cone_orient_err': nan, 'pts_bbox_NuScenes/traffic_cone_vel_err': nan, 'pts_bbox_NuScenes/traffic_cone_attr_err': nan, 'pts_bbox_NuScenes/NDS': 0.27776341220635314, 'pts_bbox_NuScenes/mAP': 0.13476739579111732}
Thanks in advance :)
from uvtr.
Hi, You can try v0.17.3 for the inference. I guess the mismatch is caused by the coordinate change in the new version. Of course, if you train it using v1.0.0rc5, it should achieve the correct results. Thanks!
from uvtr.
Related Issues (20)
- CPU memory ran out while running create_data.py HOT 2
- RuntimeError: cusolver error: CUSOLVER_STATUS_INTERNAL_ERROR, when calling `cusolverDnCreate(handle) HOT 1
- UnboundLocalError: local variable 'assign_result' referenced before assignment HOT 4
- excuese me,create data error HOT 2
- Results based on distance HOT 2
- about checkpoints
- about knowledge transfer HOT 6
- nuscene preprocess HOT 3
- Question about traning time and inference time? HOT 2
- How to compute heights Z in voxel space? HOT 2
- AttributeError: 'list' object has no attribute 'new_zeros' HOT 4
- Which mmdet3d version should I use? HOT 1
- image gtaug difference with pointaugmenting HOT 1
- Gpu memory exceeded when training a a model HOT 2
- Cuda out of memory when batch size was set to 1 HOT 2
- 'LoadPointsFromFile is not in the pipeline registry' HOT 2
- About the verison of mmdet3d HOT 1
- Prediction values becomes 'Nan' after 1 or 2 epochs HOT 1
- problem of registry
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from uvtr.