search

jingyue202205 / DSVT-AI-TRT

DSVT: Dynamic Sparse Voxel Transformer with Rotated Sets(CVPR2023),vaymo vehicle 3D Object Detection(top2), waymo cyclist 3D Object Detection(top1),waymo pedestrian 3D Object Detection(top1)
71 stars 10 forks source link

Speed up information #2

Closed d33dler closed 10 months ago

d33dler commented 1 year ago

What is the achieved speed-up with this deployment? Did you run the model without TRT and could show us the results? Thanks

LinuxCup commented 11 months ago

The author adapts the way using TensorRT API, of coruse, you need to be similar whit TensorRT. However,this method is not easy to DEBUG. I used a combination of onnx and TensorRT plugin to achieve this result。I splited the model to sub-model (dynamic-vfe, dsvt_input_layer,dsvt_block,scatter, 3Dbackbone) that running the TenosRT,and python and c++ result are the same.

python version 1_0

c++ version 1_1

LinuxCup commented 11 months ago

Here, many thanks to the author for contributions to projece deployment!!

sccdyang commented 10 months ago

The author adapts the way using TensorRT API, of coruse, you need to be similar whit TensorRT. However,this method is not easy to DEBUG. I used a combination of onnx and TensorRT plugin to achieve this result。I splited the model to sub-model (dynamic-vfe, dsvt_input_layer,dsvt_block,scatter, 3Dbackbone) that running the TenosRT,and python and c++ result are the same.

python version 1_0

c++ version 1_1

Have you compared the reasoning results between this project and the PT model to see if they are consistent? Can we open source the corresponding PT model for this project? The results of the current retraining are inconsistent.

LinuxCup commented 10 months ago

The results are consistent based on python or TensorRT environments. even if there are sight different, but effect is small.

sccdyang commented 10 months ago

The results are consistent based on python or TensorRT environments. even if there are sight different, but effect is small.

Is it your own model? I mean use jingyue202205/DSVT-AI-TRT/tools/dsvt_cbgs_dyn_pp_centerpoint.yaml to train own pt model,then get the dsvt-my.wts,then test the dsvt-my.wts as jingyue202205/DSVT-AI-TRT,finally the results between own pt and DSVT-AI-TRT of dsvt-my difference is significant。