GKT

GKT copied to clipboard

无法复现结果

8

当使用脚本训练时 python scripts/train.py +experiment=gkt_nuscenes_vehicle_kernel_7x1.yaml data.dataset_dir= data.labels_dir= 训练完毕后，指标较低  使用github上模型进行验证时 python scripts/eval.py +experiment=gkt_nuscenes_vehicle_kernel_7x1.yaml data.dataset_dir= data.labels_dir= experiment.ckptt 报错 optimizer_loop.optim_progress.optimizer.step.total.completed = self._loaded_checkpoint["global_step"] KeyError: 'global_step'

Isaac-xie

When run [eval.py] RuntimeError: unmatched '}' in format string

1

This problem occurs in torch\distributed\rendezvous.py  But in this code I can't find any '}' to solve the problem  I try to print (hostname, port, world_size, start_daemon, timeout), But...

linlion0311

我修改了gkt_nuscenes_vehicle_kernel_7x1_setting1.yaml中BEV h的尺寸，可是训练过程中，val 指标特别低，基本不变 ，不知道为什么 data: bev: # h: 400 # TODO 程序跑不通 h: 200 w: 200 # h_meters: 100.0 # TODO h_meters: 50.0 w_meters: 50.0 # python scripts/train.py +experiment=gkt_nuscenes_vehicle_kernel_7x1_setting1.yaml \...

miaoxiaoting

你好，我们在测试的时候，使用了不同的图像尺寸作为输入。但是和训练时的尺寸图像尺寸不同的时候，模型结果会差很多。请问可能的原因是什么。会不会是因为GKT的网络参数依赖于图像尺寸。

imyyf

Hi, thank you for your open source project! However, when I run eval.py , there`s an error like " KeyError:'state_dict' ", i thought it would be something wrong with 'state_dict'...

JacksonVation

Where is script/generate_labels.py file?

2

Hi, as the title written, I can only find `generate_data.py` in `GKT/segmentation/script`. Where can I find the script that generate labels? Thanks!

GoroYeh56

Dear author, where is the generate_label.py?Thx

1


emilyemliyM

Hi! Thanks a lot for the open-source project! However, I failed to reproduce the experimental results, i.e. the high FPS compared with BEVFormer. May I ask if you have made...

Zixin-Lei

How soon will the code be open sourced

4

How soon will the code be open sourced

24werewolf

zclion