Cann training camp learns the animation stylization and AOE ATC tuning of the second season of 2022 model series

1、unzip -O GBK *.zip
This command can unpack Chinese compressed packages ( Especially by windows System to linux On the system )

2、 Migration scripts
After running through

cd ..
git clone https://gitee.com/ascend/tensorflow.git
cd tensorflow/convert_tf2npu # Access kit 
pip3 install pandas
pip3 install xlrd==1.2.0
pip3 install openpyxl
pip3 install tkintertable
pip3 install google_pasta
python3 main.py -i ~/competition/LeNet

job configure as follows

Note that these two paths cannot be the bucket root
Run code

Report errors ：ImportError: cannot import name ‘tutorials’ from ‘tensorflow_core.examples’
there MNIST Data sets are used tensorflow Library to download . It needs to be changed Train.py Medium import The way .

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data

meanwhile data Modify accordingly
mnist = input_data.read_data_sets(“/home/ma-user/modelarts/user-job-dir/code/MNIST_data/”, one_hot=True)


3、 Replace the model
Download a resnet101 Model

Upload to model Folder
src/sample_process.cpp Custom code in the file . Amendment No 82 The line is as follows .
stay src/sample_process.cpp Custom code in the file . Amendment No 82 The line is as follows .

Amendment No 95~98 Documents required for behavior , Here's the picture . If it is consistent with the figure below , There is no need to modify .

4、 Animation stylization
Reference resources ：https://gitee.com/ascend/samples/tree/master/python/contrib/cartoonGAN_picture
https://www.hiascend.com/document/detail/zh/CANNCommunityEdition/51RC2alpha005/developmenttools/devtool/aoe_16_002.html

cd ~/samples/python/contrib/cartoonGAN_picture
cd model
#  To make it easy to download , Here, the original model download and model conversion commands are directly given , You can copy and execute directly . You can also refer to the above table in modelzoo Download and manually convert , For more details . 
cd ${
    HOME}/samples/python/contrib/cartoonGAN_picture/model    
wget https://modelzoo-train-atc.obs.cn-north-4.myhuaweicloud.com/003_Atc_Models/AE/ATC%20Model/cartoonization/cartoonization.pb    
wget https://modelzoo-train-atc.obs.cn-north-4.myhuaweicloud.com/003_Atc_Models/AE/ATC%20Model/cartoonization/insert_op.cfg

Use AOE tuning

 modify HwHiAiUser The user password of 
su -
passwd HwHiAiUser
apt-get install pciutils
cd ${
    HOME}/samples/python/contrib/cartoonGAN_picture/model

aoe --model="./cartoonization.pb" --output_type=FP32 --input_shape="train_real_A:1,256,256,3" --input_format=NHWC --output="cartoonizationv2_256" --framework=3 --precision_mode=allow_fp32_to_fp16 --insert_op_conf=insert_op.cfg --job_type=2

atc --model="./cartoonization.pb" --output_type=FP32 --input_shape="train_real_A:1,256,256,3" --input_format=NHWC --output="cartoonizationv2_256" --framework=3  --precision_mode=allow_fp32_to_fp16 --soc_version=Ascend310 --insert_op_conf=insert_op.cfg  


cd ~
mkdir AscendProjects
cd AscendProjects
git clone https://gitee.com/ascend/tools.git
export DDK_PATH=/home/HwHiAiUser/Ascend/ascend-toolkit/latest
export NPU_HOST_LIB=/home/HwHiAiUser/Ascend/ascend-toolkit/latest/acllib/lib64/stub
cd $HOME/AscendProjects/tools/msame/
./build.sh g++ $HOME/AscendProjects/tools/msame/out

cd ${
    HOME}/samples/python/contrib/cartoonGAN_picture/model

cd ~/AscendProjects/tools/msame/out
./msame --model ${
    HOME}/samples/python/contrib/cartoonGAN_picture/model/cartoonizationv2_256.om --output output/ --loop 100

But the tuning lasted about an hour , More than 3000 seconds





You can see our newly generated om file

recompile

cd $HOME/samples/python/contrib/cartoonGAN_picture/data
wget https://c7xcode.obs.cn-north-4.myhuaweicloud.com/models/cartoonGAN_picture/scenery.jpg
cd ../src
python3.6 cartoonization.py ../data/

In the installation numpy when , There's a problem
FileNotFoundError: [Errno 2] No such file or directory: ‘/usr/local/python3.7.5/bin/pip’

 which pip
     /home/HwHiAiUser/.local/bin/pip
 sudo ln -s /home/HwHiAiUser/.local/bin/pip /usr/local/python3.7.5/bin/pip 

 such pip Install it on python3.7.5

It's ok now

pip install --upgrade pip
 Configure the source first 
pip config set global.index-url https://pypi.tuna.tsinghua.edu.cn/simple
 Then change it to 
[global]
index-url = https://repo.huaweicloud.com/repository/pypi/simple
trusted-host = repo.huaweicloud.com
timeout = 120

pip install numpy
pip install opencv-python
find ../../../ -name "acllite_utils*"
cp ../../../common/acllite/* ./
python3.7.5 cartoonization.py ../data/

Conversion success , Go to out Let's have a look

You get the generated image .