-
### 请提出你的问题 Please ask your question
Hello,
根据https://github.com/PaddlePaddle/Paddle3D/tree/develop/docs/models/petr 下训练命令,
无论是单卡还是多卡,都无法收敛,总在某一阶段后LOSS突然增大/与Box相关的LOSS突然为0;
-
Hello,
Im trying to run the train.py file with the detr3d_res101_gridmask.py configuration but Im not able to build the Detr3D model using mmdetection3d. Do I have to do anything extra in order to …
-
### Problem Description
```
4: ---------------------------------- Results of QA Test - Tensor_hip ----------------------------------
4:
4: brightness_u8_Tensor_PKD3: FAILED
4: brightness_u8_Te…
-
Hi,
i am using this config https://github.com/megvii-research/PETR/blob/main/projects/configs/petrv2/petrv2_BEVseg.py and its respective model.
But I changed the test pipeline, with scale(twice-…
-
Hello,
Thanks for all your hard work! I saw your previous issue about your specific environment, but running into issues importing flash-attn. flash_attn_cuda isn't in the site-packages after install…
-
Will you release the code for petr-e? Or could you please share some implementation details?
Thanks
-
Hi, thanks for the great work!
I wonder how do you train the model. Can you provide the training command?
Thanks in advance!
-
Thanks for your nice work~ but I met some questions when I used this model in the customer dataset.
I trained a nice model that achieved satisfying mAP in the validation dataset and the first tes…
-
hi junjie,
In you latest work DAL, you mention a method VelocityAug to balance the velocity distribution. However, when I insert this method in bevfusion the NDS decrease, which seems different with …
liyih updated
1 month ago
-
- https://arxiv.org/abs/2104.13534
- 2021
物体検出は、コンピュータビジョンにおける基本的かつ困難なタスクであり、様々な産業アプリケーションにおいて重要な役割を果たしています。
しかし、深層学習に基づく物体検出器は、通常、より大きなストレージ要件とより長い推論時間を必要とし、その実用性を深刻に妨げています。
そのため、実用的なシナリオでは、有効性…
e4exp updated
3 years ago