英偉達環境安裝

創建虛擬環境

conda create -n paddleOCR python==3.10 -y
conda activate paddleOCRconda install jupyterlab -y
conda install ipykernel -y
python -m ipykernel install --user --name paddleOCR --display-name "paddle OCR"

下載PaddleOCR的GPU版本

下載鏈接： https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/develop/install/pip/windows-pip.html

選擇合適的版本執行命令行下載即可
然后直接使用pip下載panddleocr

pip install  paddleocr

昇騰300IDUO源碼安裝

在安裝前，先確保安裝好了昇騰的驅動和固件，并且安裝CANN
然后添加到環境變量中

export ASCEND_HOME_PATH=/usr/local/Ascend/latest
source ${ASCEND_HOME_PATH}/set_env.sh

官方安裝鏈接：https://www.hiascend.com/document/detail/zh/CANNCommunityEdition/82RC1alpha002/softwareinst/instg/instg_0001.html?Mode=PmIns&OS=Ubuntu&Software=cannToolKit

下載源碼

通過網盤分享的文件：PaddleOCR_Ascend.zip
鏈接: https://pan.baidu.com/s/1XbtJKQDoj3eKY0XRS9VMzA?pwd=e519 提取碼: e519

在本篇中，代碼下載后解壓到目錄/home/PaddleOCR_Ascend下

創建虛擬環境

# 創建虛擬環境
conda create -n paddleocr python=3.9 -y
conda activate paddleocrpip install -r requirements.txt
pip install paddlepaddle==2.6.1 paddle2onnx==1.2.4

準備推理模型

進入源碼目錄

cd /home/PaddleOCR_Ascend

創建目錄并下載模型

mkdir inference
cd inferencewget -nc https://paddleocr.bj.bcebos.com/PP-OCRv4/chinese/ch_PP-OCRv4_det_server_infer.tar
tar -xf ch_PP-OCRv4_det_server_infer.tar
wget -nc https://paddleocr.bj.bcebos.com/PP-OCRv4/chinese/ch_PP-OCRv4_rec_server_infer.tar
tar -xf ch_PP-OCRv4_rec_server_infer.tar

paddle轉onnx格式

paddle2onnx --model_dir inference/ch_PP-OCRv4_det_server_infer \--model_filename inference.pdmodel \--params_filename inference.pdiparams \--save_file inference/det/model.onnx \--opset_version 11 \--enable_onnx_checker Truepaddle2onnx --model_dir inference/ch_PP-OCRv4_rec_server_infer \--model_filename inference.pdmodel \--params_filename inference.pdiparams \--save_file inference/rec/model.onnx \--opset_version 11 \--enable_onnx_checker True

安裝相關依賴

sudo yum install mesa-libGL
pip install mindspore
pip install mindspore-lite[ascend] 

onnx轉mindir格式

cd /home/data
# 由于我的300i duo是arm架構，因此下載對應的工具
wget https://ms-release.obs.cn-north-4.myhuaweicloud.com/2.3.0rc2/MindSpore/lite/release/linux/aarch64/cloud_fusion/python37/mindspore-lite-2.3.0rc2-linux-aarch64.tar.gz
tar -zxvf mindspore-lite-2.3.0rc2-linux-aarch64.tar.gz
# 將轉換工具需要的動態鏈接庫加入環境變量LD_LIBRARY_PATH，比如我這里裝在/data/tff下
export LD_LIBRARY_PATH=/home/data/mindspore-lite-2.3.0rc2-linux-aarch64/tools/converter/lib:${LD_LIBRARY_PATH}

cd /home/data/mindspore-lite-2.3.0rc2-linux-aarch64/tools/converter/converter
./converter_lite --fmk=ONNX \--saveType=MINDIR \--optimize=ascend_oriented \--modelFile=/home/PaddleOCR_Ascend/inference/det/model.onnx \--outputFile=/home/PaddleOCR_Ascend/inference/det/model./converter_lite --fmk=ONNX \--saveType=MINDIR \--optimize=ascend_oriented \--modelFile=/home/PaddleOCR_Ascend/inference/rec/model.onnx \--outputFile=/home/PaddleOCR_Ascend/inference/rec/model

測試

python /home/PaddleOCR_Ascend/tools/infer/predict_system.py \--use_mindir=True \--gpu_id=0 \--image_dir=/home/PaddleOCR_Ascend/doc/imgs/1.jpg \--det_model_dir=/home/PaddleOCR_Ascend/inference/det/model.mindir \--rec_model_dir=/home/PaddleOCR_Ascend/inference/rec/model.mindir \--rec_char_dict_path=/home/PaddleOCR_Ascend/ppocr/utils/ppocr_keys_v1.txt \--use_angle_cls=False \--vis_font_path=/home/PaddleOCR_Ascend/doc/fonts/simfang.ttf