Kracozebr
Kracozebr
Any updates to `NonZero` operation?
Use COCO format
> Thank you so much for the contribution! I will take a look at this soon -- we need to check on the compatibility with the new torch2trt dynamic plugin...
Thanks for fast reply , no there is no such structure. Here is the code of model ``` import torch.nn as nn class LeNet5(nn.Module): def __init__(self): super().__init__() self._body = nn.Sequential(...
My last comment was deleted somehow =( Yes if I use opt_shpe_param like this: ``` opt_shape_param = [ [ [1, 1, 32, 32], # min [1, 1, 32, 32], #...
Same issue do you find the solution?
Hi, I turned out that I forgot to delete my dataset from config file, I made one more commit with fixes of config.py. What is the best way to add...
> 你可以确认一下kernel/drivers/video/rockchip/rga3目录下的代码是否有正确的被替换为更新包内的代码。 But there is no such path /drivers/video/rockchip/rga3 on OragePi5 and on Liontron Rockchip RK3588. I've tried to find directory rga3 on both devices via find on Android OS...
My vendor does not respond me) I found some linux sdk for rk3588 Is this the same one you are using? https://gitlab.com/rk3588_linux/rk/kernel/-/tree/linux-5.10/drivers/video/rockchip/rga3?ref_type=heads
Same question how to implement custom ops in rrknn-toolkit2?