swc-17/SparseDrive
SparseDrive: End-to-End Autonomous Driving via Sparse Scene Representation
PythonMIT
Issues
- 2
About coordinate
#71 opened by Zbozhou - 4
关于Ego Status from T-1输入的问题
#70 opened by guoyage - 8
训练不稳定,复现结果有差异?
#64 opened by Sunstroperao - 6
训练模块数据维度问题
#69 opened by gzspkgypt - 2
Question about: num_temp_instances for map
#68 opened by julien-seitz - 1
batch size problem
#33 opened by TmacTmac1992 - 1
Loss function問題
#59 opened by wowowolala - 2
使用sh scripts/test.sh,出现[ ] 0/3, elapsed: 0s, ETA:./tools/dist_test.sh: 第 13 行: 114146 已杀死 PYTHONPATH="$(dirname $0)/..":$PYTHONPATH python3 $(dirname "$0")/test.py $CONFIG $CHECKPOINT ${@:4}
#61 opened by shuiyihang - 6
- 3
关于预测周边车辆轨迹与自车规划轨迹的时间步长问题
#57 opened by guoyage - 3
关于使用单卡GPU训练
#53 opened by guoyage - 1
Uniad指标与官方不一致?
#54 opened by jackfenghhhhh - 7
Variation of input sampling frequency
#48 opened by julien-seitz - 2
有关双卡GPU训练报错的问题。
#63 opened by guoyage - 3
推論時間的FPS定義問題?
#65 opened by waynechu1109 - 1
有关模型验证阶段,内存out of memory问题
#66 opened by guoyage - 0
NVIDIA Driver 510 and CUDA 11.6 Compatibility Issue on RTX 4060 Ti (Also Tried 535 540)
#67 opened by wyanting314 - 6
Images Preprocess
#55 opened by waynechu1109 - 2
sh scripts/visualize.sh error
#60 opened by zhangzscn - 3
在mini数据集上使用sh scripts/kmeans.sh 问题
#58 opened by shuiyihang - 6
sh scripts/kmeans.sh run error
#23 opened by 2132660698 - 2
- 1
有关模型test时候,sparsedrive_small_stage2.py的设置问题
#52 opened by guoyage - 6
About the cam_intrinsic after data augment
#49 opened by zengwz - 1
det_head config
#50 opened by bowang9711 - 2
FLOPs calculate
#51 opened by VanLinLin - 13
stage2训练使用多线程卡死问题
#36 opened by LZH-97 - 8
How to use my own dataset?
#46 opened by suayu - 4
在自己的数据集上A10,A100, A800的表现完全不一样
#47 opened by JYS997760473 - 4
Question about the ResizeCropFlipImage?
#25 opened by Slonna - 4
- 1
如何在V100s等非Ampere架构的GPU上执行推理和训练?RuntimeError: FlashAttention only supports Ampere GPUs or newer.
#44 opened by suayu - 2
- 2
mmcv error
#34 opened by yeyueluke - 1
About the gt traj collision rate
#43 opened by JYS997760473 - 6
测试输出包含nan数据
#35 opened by 198265 - 0
使用A100 8张卡训练第一阶段loss基本都为0
#42 opened by JYS997760473 - 2
训练过程中loss变成nan了
#41 opened by suayu - 2
为什么在SparseDrive里把Sparse4D提出的denoise关掉了呢
#39 opened by JYS997760473 - 0
Calculate Flops
#40 opened by 504058568 - 4
About FlashAttention
#37 opened by Zbozhou - 3
classification loss equal 0 when training.
#30 opened by PeterJaq - 1
- 1
GPU Memory Usage and Training Time
#26 opened by zhaosonghui - 2
- 3
- 5
training with only map head failed
#32 opened by ASONG0506 - 1
Vehicle Coordinate problem
#28 opened by TmacTmac1992 - 2
loss appear to be 0 or nan during training
#27 opened by 1YuChenyao - 1
嵌入式部署问题
#22 opened by q2442590777