在Firefly AIO-3399ProC搭建rknn环境

1第一步
更改镜像源,不更改镜像源后面就会报错。
sudo vi /etc/apt/sources.list
复制下面一段，替换原来的镜像网址
deb http://wiki.t-firefly.com/firefly-rk3399-repo bionic main

See http://help.ubuntu.com/community/UpgradeNotes for how to upgrade to

newer versions of the distribution.

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic main restricted

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic main restricted

Major bug fix updates produced after the final release of the

distribution.

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic-updates main restricted

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic-updates main restricted

N.B. software from this repository is ENTIRELY UNSUPPORTED by the Ubuntu

team. Also, please note that software in universe WILL NOT receive any

review or updates from the Ubuntu security team.

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic universe

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic universe

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic-updates universe

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic-updates universe

N.B. software from this repository is ENTIRELY UNSUPPORTED by the Ubuntu

team, and may not be under a free licence. Please satisfy yourself as to

your rights to use the software. Also, please note that software in

multiverse WILL NOT receive any review or updates from the Ubuntu

security team.

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic multiverse

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic multiverse

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic-updates multiverse

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic-updates multiverse

N.B. software from this repository may not have been tested as

extensively as that contained in the main release, although it includes

newer versions of some applications which may provide useful features.

Also, please note that software in backports WILL NOT receive any review

or updates from the Ubuntu security team.

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic-backports main restricted universe multiverse

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic-backports main restricted universe multiverse

Uncomment the following two lines to add software from Canonical’s

‘partner’ repository.

This software is not part of Ubuntu, but is offered by Canonical and the

respective vendors as a service to Ubuntu users.

deb http://archive.canonical.com/ubuntu bionic partner

deb-src http://archive.canonical.com/ubuntu bionic partner

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic-security main restricted

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic-security main restricted

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic-security universe

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic-security universe

deb http://mirrors.aliyun.com/ubuntu-ports/ bionic-security multiverse

deb-src http://mirrors.aliyun.com/ubuntu-ports/ bionic-security multiverse

然后更新源
sudo apt-get update
2.1 安装python3.7
sudo apt update #检查可更新文件
sudo apt install software-properties-common #安装可添加源的工具
sudo add-apt-repository ppa:deadsnakes/ppa #添加源，否则会无法找到python3.7-dev软件包
sudo apt install python3.7-dev #安装python3.7
2.2 1.2修改python默认为python3.7
为了因为C++调用python时，默认是调用python2.7，这导致很多python3.7的语法报错。所以需要更改python默认软连接
查看路径python,python3.7路径
which python python3.7
/usr/bin/python
/usr/bin/python3.7
sudo rm /usr/bin/python
sudo rm /usr/bin/python3
创建python3.软连接到python&Python3
sudo ln -s /usr/bin/python3.7 /usr/bin/python
sudo ln -s /usr/bin/python3.7 /usr/bin/python3
3 sudo apt-get update 更新包和安装依赖包
出现报错： No module named 'apt_pkg’
进行下面操作
sudo vi /usr/bin/apt-add-repository
#! /usr/bin/python3.6
cd /usr/lib/python3/dist-packages
sudo cp apt_pkg.cpython-36m-aarch64-linux-gnu.so apt_pkg.so
然后 sudo apt-get update
接下来安装依赖的包
pip3 install Cython
sudo apt-get install gcc python3-dev
sudo apt-get install libhdf5-dev
sudo apt-get install cmake gcc g++ libprotobuf-dev protobuf-compiler libgfortran5-dbg libopenblas-dev gfortran libprotoc-dev
4 安装下面的包
pip3 install wheel opencv-python4.6.0.66 -i https://pypi.tuna.tsinghua.edu.cn/simple
pip3 install -i https://pypi.tuna.tsinghua.edu.cn/simple numpy1.20.1

sudo apt-get install libhdf5-dev
rknn_toolkit-1.7.1-cp37-cp37m-linux_aarch64.whl放到目录下面，提前安装,不依赖包，编译的速度太慢，而且容易报错，这样大幅度提高了安装效率。
pip3 install --no-deps -i https://pypi.tuna.tsinghua.edu.cn/simple rknn_toolkit-1.7.1-cp37-cp37m-linux_aarch64.whl
pip3 install h5py2.8.0 -i https://pypi.tuna.tsinghua.edu.cn/simple #安装h5py包，编译完成即可进行下一步
pip3 install scipy1.7.0 -i https://pypi.tuna.tsinghua.edu.cn/simple #安装scipy包，编译完成即可进行下一步
这时候提示缺少了这些包，一一去安装上。
ERROR: pip’s dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
rknn-toolkit 1.7.1 requires dill0.2.8.2, which is not installed.
rknn-toolkit 1.7.1 requires flatbuffers1.10, which is not installed.
rknn-toolkit 1.7.1 requires lmdb0.93, which is not installed.
rknn-toolkit 1.7.1 requires networkx1.11, which is not installed.
rknn-toolkit 1.7.1 requires onnx-tf1.2.1, which is not installed.
rknn-toolkit 1.7.1 requires ply3.11, which is not installed.
rknn-toolkit 1.7.1 requires psutil5.6.2, which is not installed.
rknn-toolkit 1.7.1 requires ruamel.yaml0.15.81, which is not installed.
rknn-toolkit 1.7.1 requires sklearn0.0, which is not installed.
rknn-toolkit 1.7.1 requires flask1.0.2, but you have flask 2.2.2 which is incompatible.
rknn-toolkit 1.7.1 requires numpy1.16.3, but you have numpy 1.20.0 which is incompatible.
rknn-toolkit 1.7.1 requires Pillow5.3.0, but you have pillow 5.1.0 which is incompatible.
rknn-toolkit 1.7.1 requires protobuf3.11.2, but you have protobuf 4.21.5 which is incompatible.
rknn-toolkit 1.7.1 requires requests2.22.0, but you have requests 2.18.4 which is incompatible.
安装缺少的包：

pip3 install -i https://pypi.tuna.tsinghua.edu.cn/simple dill0.2.8.2 flatbuffers1.10 lmdb0.93 networkx1.11 onnx-tf1.2.1 ply3.11 psutil5.6.2 ruamel.yaml0.15.81 sklearn0.0 protobuf3.11.2 onnx==1.6.0
将 下载好的将下载好的tensorflow-1.14.0-cp37-none-linux_aarch64.whl 放置目录下
然后安装 pip3 install -i https://pypi.tuna.tsinghua.edu.cn/simple tensorflow-1.14.0-cp37-none-linux_aarch64.whl
5 更新rknn驱动：sudo apt install firefly-3399pronpu-driver
6 运行tensorflow的案例程序报错，将"from tensorflow.contrib import cloud"去掉：
ImportError: cannot import name ‘cloud’ from ‘tensorflow.contrib’ (/home/firefly/.local/lib/python3.7/site-packages/tensorflow/contrib/init.py)
E Only support ntb mode on ARM64 platform. But can not find device with ntb mode.
E Catch exception when init runtime!
E Traceback (most recent call last):
E File “rknn/api/rknn_base.py”, line 868, in rknn.api.rknn_base.RKNNBase.init_runtime
E File “rknn/api/rknn_runtime.py”, line 171, in rknn.api.rknn_runtime.RKNNRuntime.init
E File “rknn/api/rknn_platform_utils.py”, line 307, in rknn.api.rknn_platform_utils.start_ntp_or_adb
E Exception: Init runtime environment failed!
E Please feedback the detailed log file to the RKNN Toolkit development team.
E You can also check github issues: https://github.com/rockchip-linux/rknn-toolkit/issues
7 参考的博客
6.1 https://blog.csdn.net/Lxbbdxg/article/details/124451477
6.2 https://blog.csdn.net/HTS514286/article/details/123644488