Tengine
Tengine copied to clipboard
OAID
Tengine is a lite, high performance, modular inference engine for embedded device
编译cuda版本，用gpu去跑官方给的yolov3和yolov4模型，跑一张图片，把-r设成100，发现内存吃不消了。 而且跑nanodnet在跑到prerun_graph_multithread(graph, opt)这一步的时候也会报错，好像是你们自定义的vector访问越界了
I tried to use Tengine on M1 MacBook Air. But, I could not build Tengine. ### System information (version) - Tengine => Tengine Lite release v1.2 - Operating System /...
Need to run custom mobilenetV2 based "*.tflite" model on the Khadas-VIM3's NPU. Tried converting a custom mobilenetV2 based classifier model from "*.tflite" to "*.tmfile", but faced the following error. ...
在板子上使用模型进行推理时速度很快，但j加载模型初始化时间很久，大概十几秒，能否有某种预编译模型的方式来把初始化速度降低到1秒以下呢，谢谢
代码执行if (prerun_graph_multithread(graph, opt) < 0)时在rv1126板子上用时很久大概23秒，请问该怎么解决呢，谢谢
用risc-v交叉编译工具链交叉编译tengine时，会出现如下问题 Tengine/source/device/cpu/op/conv/risc-v/lp64dv/im2col_fp32_1x1.S: Assembler messages: Tengine/source/device/cpu/op/conv/risc-v/lp64dv/im2col_fp32_1x1.S:60: Error: unrecognized opcode `vsetvli t0,a0,e32' Tengine/source/device/cpu/op/conv/risc-v/lp64dv/im2col_fp32_1x1.S:82: Error: unrecognized opcode `vlw.v v0,(t3)' Tengine/source/device/cpu/op/conv/risc-v/lp64dv/im2col_fp32_1x1.S:83: Error: unrecognized opcode `vlw.v v1,(t1)' Tengine/source/device/cpu/op/conv/risc-v/lp64dv/im2col_fp32_1x1.S:87: Error: unrecognized opcode `vsw.v v0,(a2)' Tengine/source/device/cpu/op/conv/risc-v/lp64dv/im2col_fp32_1x1.S:89: Error:...
squeezenet_v1.1的模型和数据来自ncnn的github，测试图片cat.jpg来自 tengine example 转换 ncnn/examples 里的 squeezenet_v1.1.param convert_tool -f ncnn -p squeezenet_v1.1.param -m squeezenet_v1.1.bin -o squeezenet_v1.1.tmfile 测试结果完全不对 tm_classification -m squeezenet_v1.1_from_ncnn_new.tmfile -i images/cat.jpg -g 227,227 -w 104.0,117.0,123.0 -s 1.0,1.0,1.0 tengine-lite library...
