-
root@localhost:/opt/FastDeploy-develop/build# cmake .. -DENABLE_ORT_BACKEND=ON \
> -DENABLE_RKNPU2_BACKEND=ON \
> -DENABLE_VISION=ON \
> -DRKNN2_TARGET_SOC=RK356X \
> …
-
We all would love to see only implementations compatible with the soon to be ratified 1.0 standard. This would ease the software burden and require Linux (etc) distros to have to deal with only RV64GC…
-
## 环境
- 【FastDeploy版本】： 当前github主线 (rev. b672e32daedb2d539984a43ac55d16fb6dde6234)
- 【编译命令】
```
cmake .. -DENABLE_ORT_BACKEND=ON \
-DENABLE_OPENVINO_BACKEND=ON \
-DCMAKE_INST…
-
模型转换过程中
===group_idx: 104move tensor op_168_0_Add_f32 from timestep 1 to timestep 7merge timestep 2 to timestep 1
merge timestep 3 to timestep 2
merge timestep 3 to timestep 2
merge timestep 3 to …
-
自己编译坑太多了，官方能不能给一个编译好的库？
我现在编译好了后，执行推理onnx模型的时候，报如下错误
File "/home/deepnature/anaconda3/envs/paddle2.4/lib/python3.7/site-packages/fastdeploy/c_lib_wrap.py", line 164, in
from .libs.fastdeploy…
-
## 环境
- 【FastDeploy版本】： 1.0.4
- 【编译命令】
```
cd FastDeploy/python
export ENABLE_ORT_BACKEND=ON
export ENABLE_PADDLE_BACKEND=ON
export ENABLE_OPENVINO_BACKEND=ON
export ENABLE_VISION=ON
export…
-
### 运行环境
软件
```
使用了readme中提供的官方sophon docker 镜像
```
硬件
SC7 TPU 算力卡， 由于散热问题目前没有安装到服务器中，仅以此项目作软件模型编译测试
### 问题描述：
根据 readme 中的 instruction 进行到模型编译一步时:
```
./compile.sh
```
具体命令是：
``…
-
-
同題目，想請教說目前compiler是否有支持pytorch QAT後輸出的onnx graph?
-
It seems that fip.bin contains different pieces of software with different licenses:
* ARM Trusted Firmware (BSD-3-Clause license)
* U-Boot 2022.10 (GPL v2 license)
The Gnu Public License of U-…