Pinned issues
Issues
- 1
- 0
- 2
关于self.num_points和self.num_levels
#73 opened by kingofstu - 3
- 4
如何生成nuscenes_info_train.pkl文件?
#72 opened by kingofstu - 0
cu文件用什么编译器编写的,怎么调试啊
#71 opened by rubbish001 - 0
牛,训练11天,还是a100,普通人还能跑吗
#68 opened by rubbish001 - 1
What's the _delete_ parameter means?
#70 opened by ZitengXue - 1
decoder layer里的add & norm
#69 opened by Azeral0619 - 1
pretraining SparseBEV with only 1 frame
#67 opened by jeff770 - 2
sampling_4d
#66 opened by jhjun724 - 8
SparseOcc当中sparse decoder部分的最后一层query数量是65536吗?以及第一部分的体素化分辨率是1.6m*1.6m*1.6m吗?
#61 opened by cheng-peng-git - 3
sampling points lidar2img问题
#64 opened by zhaoyangwei123 - 3
Performance of sparse voxel decoder
#65 opened by Yzichen - 6
- 3
About max points in msmv_sampling.cpp
#63 opened by zbaishancha - 5
SparseOcc questions
#47 opened by vehxianfish - 6
self.init_query_bbox = nn.Embedding(self.num_query, 10) . This parameter doesn't seem to have been updated.
#60 opened by zbaishancha - 1
- 3
- 8
Using flip augmentation in the training, how to align the preds bbox(on the flip imgs) to the unflip gt bbox of original ann_file?
#56 opened by Slonna - 2
- 1
Online Inference
#51 opened by YoushaaMurhij - 3
- 5
Hi, I want to know that what isthe gen_sweep_info.py used for? Does SparseBEV use nuscenes sweeps data to train and inference?
#53 opened by Slonna - 1
提Issue的各位哥哥姐姐们能不能先点个star啊
#54 opened by afterthat97 - 2
It seems that use the split val dataset for training besides the original split trainset under the nuscenes leaderboard submission config setting ?
#50 opened by Slonna - 1
- 4
Evaluation Error
#48 opened by smritish21 - 4
关于预训练权重下载问题
#32 opened by BarryGustin - 2
- 2
Is there ablation study about the train_interval?
#44 opened by tang-hy - 6
预训练模型下载链接(国内网盘)
#36 opened by CeoiZidung - 2
Where in the code do you handle ego motion?
#43 opened by Slonna - 2
Question about absolute velo
#42 opened by iMbaGong - 1
CUDA out of menory when inference
#27 opened by Orange-rightperson - 2
simple_test_{online, offline} difference
#41 opened by Lewis-Lu - 5
img_curr_large = img[:, i] # [B, 6, C, H, W] IndexError: index 1 is out of bounds for dimension 1 with size 1
#40 opened by matrix97317 - 2
V2-99 和Dual-Swin-T哪个主干网络好
#34 opened by rubbish001 - 2
Cuda out of memory
#39 opened by Orange-rightperson - 0
关于三个模块
#38 opened by GreatZeZuo - 2
Can't find `sample_infos = pickle.load(open(os.path.join(args.data_root, 'nuscenes_infos_train.pkl'), 'rb'))`
#37 opened by matrix97317 - 2
- 5
CUDA out of memory
#29 opened by Orange-rightperson - 1
Deployment questions
#33 opened by BarathwajAnandan - 1
About the generation of training data
#31 opened by tang-hy - 5
Question about more frames
#30 opened by Debrove - 2
- 8
visualize
#25 opened by zhujun3753 - 3