SA-SSD

SA-SSD copied to clipboard

Hi! I recently built the environment and successfully started training. Is there a way to start the training from a checkpoint / saved weights? I did not see this option...

BrandonBian

we can achieve the 3d mid to 84.50. but the bev mid still around 90.0, what wrong with me

3

your's Car [email protected], 0.70, 0.70: bev AP:96.55, 92.79, 90.32 3d AP:93.13, 84.54, 81.71 mine Car [email protected], 0.70, 0.70: bev AP:96.51, 90.37, 87.76 we want to know, you have split the...

yangyongguang

RuntimeError:all input tensors must be on the same device. received cuda:0 and cpu

1

my pytorch version is 1.7.1 cuda version = 11.1 (because of RTX 3090) when run `python train.py ../configs/car_cfg.py` RuntimEerror:all input tensors must be on the same device. received cuda:0 and...

june2678

How to improve validation mAP using multi-GPUs

11

I trained the model with 8GPUs, with your defaut hyper-parameters, however, the mAP is lower than single GPU. Car [email protected], 0.70, 0.70: bbox AP:95.64, 82.39, 77.33 bev AP:92.46, 79.22, 72.20...

zyms5244

I use multi-GPU trainning,but errors occurs: ``` Traceback (most recent call last): File "./train.py", line 131, in main() File "./train.py", line 82, in main model = MMDistributedDataParallel(model.cuda(),find_unused_parameters=True) File "/home/ubuntu-502/xu/CIA-SSD/envs/CIA-SSD/lib/python3.6/site-packages/torch/nn/parallel/distributed.py", line...

vehxianfish

Result reproduction in update version.

5

Hi, I notice this repo is updated in the training process and optimization. Use the previous commit, I can reproduce the 84.3 in Car [email protected]. However, the performance in the...

SlingHe

command 'gcc' failed

1

/ATen/core/TensorBody.h:341:7: note: declared here T * data() const { ^ error: command 'gcc' failed with exit status 1

cuge1995

(ssd) root@logic:~/SA-SSD/tools# python ./test.py ../configs/car_cfg.py ../20200701/checkpoint_epoch_80.pth [40, 1600, 1408] Evaluate on ['Car'] ==> Loading parameters from checkpoint ../20200701/checkpoint_epoch_80.pth to GPU ==> Checkpoint trained from version: none ==> Done (loaded 148/148)...

sherlockzym

(ssd) root@logic:~/SA-SSD/tools# python train.py ../configs/car_cfg.py 2021-02-26 09:22:00,718 - INFO - Distributed training: False 2021-02-26 09:22:00,718 - INFO - Set random seed to 0 [40, 1600, 1408] load 14357 Car database...

sherlockzym

raise RuntimeError(MISSING_LIBDEVICE_FILE_MSG.format(arch=arch)) RuntimeError: Missing libdevice file for compute_50. Please ensure you have package cudatoolkit 7.5. Install package by: conda install cudatoolkit=7.5

1

(ssd) root@logic:~/SA-SSD# python tools/create_data.py Traceback (most recent call last): File "tools/create_data.py", line 8, in from mmdet.core.bbox3d.geometry import remove_outside_points, points_in_rbbox, box_camera_to_lidar File "/root/SA-SSD/mmdet/core/__init__.py", line 1, in from .anchor import * #...

sherlockzym