Qingwen Zhang comments

Results 109 comments of


                                            Qingwen Zhang

faiss::gpu::runMatrixMult ... cublas failed (13): (1024, 12) x (256, 12)' = (1024, 256) gemm params m 256 n 1024 k 12 trA T trB N lda 12 ldb 12 ldc 256

- env: python==3.9, cuda==11.4, faiss-gpu==1.7.4 / A100 Met also in the env above, but haven't tried the solution to downgrade the python version. --- After downgrading the python version to...

缩进 and 显示有问题

同步目录问题已在最新版作者的更新后正常显示 --- 但是缩进的问题还是存在

缩进 and 显示有问题

原来如此，谢谢！ @wangyang0210 能多问一句，在`博客园本身再编辑处理` 这步操作具体是什么步骤呀

缩进 and 显示有问题

奥看懂了谢谢~ 那我关闭啦

[Registration] PCL-NDT codes calculate gauss_c1, c2 may have problem? equation 6.7

Sorry, not really #5056 is about the gauss_c2, and it discussed the outlier ratio and c2. But My question is **more related to the c1 and the formulation with codes...

[gmapping-5] process has died

> Hi guy, we just met the same problem and we were confused about it. And do you fixt it? if so , could you provide some suggestion. thnaks! sorry,...

cuda-pcl source

@Haoyu-NV 1 year later, Any plan?

Source code missing?

Artifacts on Reconstruction (KITTI)

the default config in ROS wrap launch, enable_icp=true and about the align one, I think we must use the `minkidir` for voxblox which has a totally different inverse log etc,...

[Q] Using WandB Sweep + SLURM + Pytorch Lightning DDP + Multiple GPUs

> my first instinct would be to only use a single task and use `Trainer(accelerator="gpu", devices=4, strategy="ddp")` in your train.py to spin up the parallel GPU processes. You will need...