darrenearl

RuntimeError: Only tuples, lists and Variables are supported as JIT inputs/outputs. Dictionaries and strings are also accepted, but their usage is not recommended. Here, received an input of unsupported type: int

3

(pcdet) exinova@exinova-B560M-AORUS-PRO-AX:/media/exinova/ssd/CUDA-PointPillars/tool$ python exporter.py --ckpt pointpillar_7728.pth --cfg_file cfgs/custom_models/pointpillar.yaml 2022-10-28 14:55:09,684 INFO ------ Convert OpenPCDet model for TensorRT ------ 2022-10-28 14:55:11,488 INFO ==> Loading parameters from checkpoint pointpillar_7728.pth to CPU 2022-10-28...

How do I train myself dataset?

1

hi, I want to train my self dataset with centerpoint, but it's a little dificult to read the WaymoDataset. My dataset only has lidars data(x, y, z, r), and the...

KeyError: 'CenterHead is not in the head registry'

3

(centerpoint) exinova@exinova-B560M-AORUS-PRO-AX:/media/exinova/ssd/CenterPoint-master_py$ python tools/train.py configs/nusc/pp/nusc_centerpoint_pp_02voxel_two_pfn_10sweep.py 2022-12-15 16:35:21,463 - INFO - Distributed training: False 2022-12-15 16:35:21,463 - INFO - torch.backends.cudnn.benchmark: False 2022-12-15 16:35:21,484 - INFO - Finish RPN Initialization Traceback (most...

关于qwen2-1.5b模型的问题

5

作者你好，我在用qwen2-1.5b模型的时候，量化成功，但是跑出来结果不对，请问量化参数如何配置呢？

作者你好，我在看您的Marlin w4a8代码中，fetch_to_registers函数中有一句： `int4* sh_s3_stage = sh_s3 + s3_sh_stage * ((group_blocks / thread_k_blocks) * (pipe / (group_blocks / thread_k_blocks)));` 这行代码，我发现pipe=0和1的时候，偏移为0，pipe=2和3的时候，偏移为64，这里为什么是64而不是32呢？ 我令m = 16,k = 512, n = 256, group_size = 128,设置矩阵乘法thread_k...

关于group_size的问题

1

你好，我在跑您得代码得时候，修改了一下group_size=16,但是最终结果会有nan和inf值，请问需要修改核函数哪些地方呢？