Amireux52

swinT_example.py swin_base_patch4_window7_224

1

hello,there is a sad：model = timm.create_model('swin_base_patch4_window7_224', pretrained=True) in 85 line of swinT_example.py, and how i can get 'swin_base_patch4_window7_224'?Looking forword to your reply! thanks

ERROR: The function received no value for the required argument: img_pattern

4

ERROR: The function received no value for the required argument: img_pattern Usage: predict.py IMG_PATTERN optional flags: --mask_pattern | --weights_path | --out_dir | --side_by_side | --video For detailed information on this...

how to visualize Figure 5,Figure 6, Figure7

4

Interested in your paper, how to visualize Figure 5,Figure 6, Figure7？Thanks

IndexError: list index out of range

4

when i run python tools/eval.py -n yolox-s -c yolox_s.pth -b 64 -d 8 --conf 0.001 [--fp16] [--fuse] ,the problem of IndexError: list index out of range occured, help,thanks

KeyError: 'Petr3D_seg: "PETRHead_seg: \'SinePositionalEncoding3D is not in the position encoding registry\'"'

3

When i try to add the segmentation task from PETRv2 to StreamPETR, i put【petrv2_ BEVseg.py, petr3d_Seg. py, petr_head_seg.py】 in the corresponding location under the StreamPETR project, i also add 【from...

ModuleNotFoundError: No module named 'roi_align',

2

ModuleNotFoundError: No module named 'roi_align',请问这个解决的详细步骤，谢谢

code about ReasonNet

7

作者您好，请问ReasonNet的代码以及环境配置可以开源分享吗，殷切盼复呢，我的邮箱[email protected]

老师，执行python tools/create_infos_av2/create_av2_infos.py命令，无法生成av2_train_infos.pkl，出现了如下问题：  请老师教我怎么解决这个问题，感谢，盼复

关于论文名Exploring Object-Centric Temporal Modeling for Efficient Multi-View 3D Object Detection

3

您好，请问为什么是以Object-Centric的目标检测方法呢，文中将其与BEV时序以及perspective时序进行了比较，不过没有琢磨明白什么叫做以Object-Centric的目标检测呢？盼复，谢谢

projects/configs/StreamPETR/stream_petr_vov_flash_800_bs2_seq_24e.py文件里用的是MultiHeadAttention，但是projects/mmdet3d_plugin/models/utils/petr_transformer.py里面没有MultiHeadAttention，只有PETRMultiheadAttention

9

老师您好，我发现projects/configs/StreamPETR/stream_petr_vov_flash_800_bs2_seq_24e.py文件里用的是MultiHeadAttention,如下图所示：  但是projects/mmdet3d_plugin/models/utils/petr_transformer.py里面没有MultiHeadAttention，只有PETRMultiheadAttention，如下图：  但是代码仍然可以正常跑起来，但是在debug的时候，只能进入PETRMultiheadFlashAttention，没有进入MultiheadAttention（我debug几次，没有看到），如图：  于是我将projects/mmdet3d_plugin/models/utils/petr_transformer.py中的PETRTemporalTransformer进行实例化：我将projects/configs/StreamPETR/stream_petr_vov_flash_800_bs2_seq_24e.py中关于PETRTemporalTransformer的信息粘贴到了projects/mmdet3d_plugin/models/utils/petr_transformer.py中的class PETRTemporalTransformer(BaseModule)模块的结尾，如图：   仍然可以打印MultiheadAttention如图：  请问：1、 MultiHeadAttention和PETRMultiheadFlashAttention是同一个东西吗 2、 MultiHeadAttention和PETRMultiheadAttention在网络中分别起到什么作用 盼复！！！感谢老师