FengmingGo

在Windows系统下编译Cuda引擎时报错

2

具体是从这开始报错的： C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.4\include\cuda_fp16.hpp(1973): error: asm operand type size(8) does not match type/size implied by constraint 'r' asm ("ld.global.nc.b32 %0, [%1];" : "=r"(*(reinterpret_cast(&(ret)))) : "r"(ptr)); ^ C:\Program Files\NVIDIA...

C2664“cudaError_t cudaMalloc(void **,size_t)”: 无法将参数 1 从“float **”转换为“void ** KataGo\cpp\neuralnet\cudabackend.cpp

1

Just compiled a cuda backend engine for windows. However, this problem occurs in lines 2099, 2101, 2106, 2108, 2111, 2114, 2878, 2879 of cudabackend.cpp. Of course I didn't make any...

Report an error

3

  I just cloned the latest source code to compile an engine that uses the TensorRT(plan cache)backend, but I got this error. What is the reason for this?

About TensorRT 10

2

 All the .dll and .lib files in the TensorRT 10 zip package for Windows versions are suffixed with "_10", which can lead to an issue where libnvinfer.lib cannot be...

Report a typo

2

logging.info("Shuffled data path does not exist, there seems to be no data or not enough data yet, qutting: %s" % curdatadir) logging.info("Shuffled data train.json file does not exist, there seems...

Presenting a problem that arose while exporting model weights, accompanied by the corresponding resolution

5

## During the process of exporting model weights today, the following error was encountered: PS E:\selfplay\KataGo\python> python ./export_model_pytorch.py -checkpoint "E:\selfplay\train\checkpoint.ckpt" -export-dir E:\selfplay\models -filename-prefix b1c6nbt -model-name b1c6nbt ['./export_model_pytorch.py', '-checkpoint', 'E:\\selfplay\\train\\checkpoint.ckpt', '-export-dir',...