Issues
- 4
请问可以在ubuntu16.04+cuda9.0上运行吗?编译遇到fp16的错误
#19 opened by zxt881108 - 1
亲,DCNv2中有些地方不太明白,有人帮忙解答一下吗?
#62 opened by liyuyuan6969 - 1
- 0
- 0
关于./lean/onnx/include的配置
#60 opened by YQMCU - 1
我下载了lean.zip,但是下载无法下载.onnx,有谁给我发一下呀,我把包发出来。
#58 opened by zsffuture - 0
您好,编译的时候出现 未定义的引用,请问是什么原因导致的?
#59 opened by long2double - 0
CMakeLists.txt编译版本有吗
#57 opened by chen1234520 - 1
- 1
大佬,windows的依赖库lean.zip下载地址挂了,有新的下载地址吗
#55 opened by whiteQ3Q - 1
fairmot
#54 opened by xjsxujingsong - 0
yolov5 detect gray image
#48 opened by lyxbyr - 0
你好作者,我序列化ctdet_coco_dla_2x.onnx到trt文件了,反序列化的时需要那些步骤,一值不成功。我使用的是tensorrt-onnx7.0,tensorrt7.0。
#53 opened by scuizhibin - 0
RuntimeError: Attempted to trace DCNv2Function, but tracing of legacy functions is not supported
#52 opened by scuizhibin - 0
大佬,这个可以在linux上实现吗?
#51 opened by scuizhibin - 0
合作交流
#50 opened by mymutouren - 0
大佬 小白看不懂 求助 怎样trt加速pb。。
#49 opened by henbucuoshanghai - 0
creator->createPlugin(name.c_str()返回null
#47 opened by gentlebreeze1 - 3
- 0
src/caffeplugin/plugins/ChannelMultiplicationLayer.cu(14): error: more than one conversion function from "const TRTInfer::halfloat" to a built-in type applies:
#46 opened by wolfworld6 - 4
- 2
libprotobuf FATAL
#36 opened by StrongerzZ - 0
coco_tracking.onnx模型可以转换为trt模型推理吗?
#45 opened by shliang0603 - 0
- 0
onnx模型无法下载
#43 opened by hufangjian - 10
ubuntu make error
#42 opened by gentlebreeze1 - 2
关于plugin问题enqueue问题?
#41 opened by linhezheng19 - 2
- 2
pytorch1.5 无法使用plugin_onddnx_export.py 导出模型
#20 opened by justinyym - 2
请问无法解析_std_reverse_trivially_8 外部符号 这错误怎么解决呢?
#40 opened by scuizhibin - 1
那个resize(2)是啥意思
#29 opened by manhongnie - 1
可以把自定义的resnet50的网络转成trt模型吗?
#38 opened by jxuepeng - 5
在 Ubuntu 下面怎么编译啊
#33 opened by Qjizhi - 0
- 7
ubuntu 16.04下用Makefile编译
#16 opened by EthanGuan - 1
CMakeList编译
#17 opened by lyxlty - 1
cmakelist.txt
#32 opened by skeeker - 1
这个是基于tensorrt7吗
#24 opened by shaominghe - 0
fp16的相关问题?如何设置fp16前向推理?
#31 opened by BaofengZan - 1
添加cuda实现
#28 opened by manhongnie - 2
- 1
集成libtorch出错
#26 opened by bigpo - 2
怎样才能支持转INT 8或者INT16?
#25 opened by lyxbyr - 2
请问可以改成使用python调用tensorrt吗
#21 opened by 17702513221 - 2
对于tensorrt不支持的操作有哪些解决方案?
#12 opened by Qjizhi - 1
- 1
- 4
- 4
想了解一下register_custom_op_symbolic自定义运算符的原理
#14 opened by 13213085 - 1