mirrors
/
PaddleClas
mirror of https://github.com/PaddlePaddle/PaddleClas.git
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

Fix hyperlink and support arm7 (#670) 

* Update hyperlink, test=document_fix

* Fix to support arm7
pull/674/head
Tingquan Gao 2021-04-09 16:21:07 +08:00 committed by GitHub
parent a6e2114e32
commit 97f4f557e5
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 36 additions and 34 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

33
deploy/lite/Makefile
View File

@ -9,20 +9,27 @@ THIRD_PARTY_DIR=${LITE_ROOT}/third_party
OPENCV_VERSION=opencv4.1.0
OPENCV_LIBS = ${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/libs/libopencv_imgcodecs.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/libs/libopencv_imgproc.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/libs/libopencv_core.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/libtegra_hal.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/liblibjpeg-turbo.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/liblibwebp.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/liblibpng.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/liblibjasper.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/liblibtiff.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/libIlmImf.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/libtbb.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/arm64-v8a/3rdparty/libs/libcpufeatures.a
ifeq (${ARM_ABI}, arm8)
ARM_PATH=arm64-v8a
endif
ifeq (${ARM_ABI}, arm7)
ARM_PATH=armeabi-v7a
endif
OPENCV_INCLUDE = -I../../../third_party/${OPENCV_VERSION}/arm64-v8a/include
OPENCV_LIBS = ${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/libs/libopencv_imgcodecs.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/libs/libopencv_imgproc.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/libs/libopencv_core.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/libtegra_hal.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/liblibjpeg-turbo.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/liblibwebp.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/liblibpng.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/liblibjasper.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/liblibtiff.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/libIlmImf.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/libtbb.a \
${THIRD_PARTY_DIR}/${OPENCV_VERSION}/${ARM_PATH}/3rdparty/libs/libcpufeatures.a
OPENCV_INCLUDE = -I../../../third_party/${OPENCV_VERSION}/${ARM_PATH}/include
CXX_INCLUDES = $(INCLUDES) ${OPENCV_INCLUDE} -I$(LITE_ROOT)/cxx/include

19
deploy/lite/readme.md
View File

@ -25,8 +25,8 @@ Paddle Lite是飞桨轻量化推理引擎，为手机、IOT端提供高效推理
1. [建议]直接下载，预测库下载链接如下：
|平台|预测库下载链接|
|-|-|
|Android|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/Android/inference_lite_lib.android.armv7.gcc.c++_static.with_extra.CV_ON.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/Android/inference_lite_lib.android.armv8.gcc.c++_static.with_extra.CV_ON.tar.gz)|
|iOS|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/iOS/inference_lite_lib.ios.armv7.with_extra.CV_ON.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/iOS/inference_lite_lib.ios64.armv8.with_extra.CV_ON.tar.gz)|
|Android|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/Android/gcc/inference_lite_lib.android.armv7.gcc.c++_static.with_extra.with_cv.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/Android/gcc/inference_lite_lib.android.armv8.gcc.c++_static.with_extra.with_cv.tar.gz)|
|iOS|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/iOS/inference_lite_lib.ios.armv7.with_cv.with_extra.tiny_publish.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/iOS/inference_lite_lib.ios.armv8.with_cv.with_extra.tiny_publish.tar.gz)|
**注**：
1. 如果是从 Paddle-Lite [官方文档](https://paddle-lite.readthedocs.io/zh/latest/quick_start/release_lib.html#android-toolchain-gcc)下载的预测库，
@ -83,9 +83,10 @@ Paddle-Lite 提供了多种策略来自动优化原始的模型，其中包括
#### 2.1.1 [建议]pip安装paddlelite并进行转换
Python下安装 `paddlelite`，目前最高支持`Python3.7`。
**注意**：`paddlelite`whl包版本必须和预测库版本对应。
```shell
pip install paddlelite
pip install paddlelite==2.8
```
之后使用`paddle_lite_opt`工具可以进行inference模型的转换。`paddle_lite_opt`的部分参数如下
@ -132,17 +133,13 @@ cd build.opt/lite/api/
```shell
# 进入PaddleClas根目录
cd PaddleClas_root_path
export PYTHONPATH=$PWD
# 下载并解压预训练模型
wget https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_large_x1_0_pretrained.tar
tar -xf MobileNetV3_large_x1_0_pretrained.tar
# 将预训练模型导出为inference模型
python tools/export_model.py -m MobileNetV3_large_x1_0 -p ./MobileNetV3_large_x1_0_pretrained/ -o ./MobileNetV3_large_x1_0_inference/
# 下载并解压inference模型
wget https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/MobileNetV3_large_x1_0_infer.tar
tar -xf MobileNetV3_large_x1_0_infer.tar
# 将inference模型转化为Paddle-Lite优化模型
paddle_lite_opt --model_file=./MobileNetV3_large_x1_0_inference/model --param_file=./MobileNetV3_large_x1_0_inference/params --optimize_out=./MobileNetV3_large_x1_0
paddle_lite_opt --model_file=./MobileNetV3_large_x1_0_infer/inference.pdmodel --param_file=./MobileNetV3_large_x1_0_infer/inference.pdiparams --optimize_out=./MobileNetV3_large_x1_0
```
最终在当前文件夹下生成`MobileNetV3_large_x1_0.nb`的文件。

18
deploy/lite/readme_en.md
View File

@ -26,8 +26,8 @@ For the detailed compilation directions of different development environments, p
|Platform|Inference Library Download Link|
|-|-|
|Android|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/Android/inference_lite_lib.android.armv7.gcc.c++_static.with_extra.CV_ON.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/Android/inference_lite_lib.android.armv8.gcc.c++_static.with_extra.CV_ON.tar.gz)|
|iOS|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/iOS/inference_lite_lib.ios.armv7.with_extra.CV_ON.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.6.1/iOS/inference_lite_lib.ios64.armv8.with_extra.CV_ON.tar.gz)|
|Android|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/Android/gcc/inference_lite_lib.android.armv7.gcc.c++_static.with_extra.with_cv.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/Android/gcc/inference_lite_lib.android.armv8.gcc.c++_static.with_extra.with_cv.tar.gz)|
|iOS|[arm7](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/iOS/inference_lite_lib.ios.armv7.with_cv.with_extra.tiny_publish.tar.gz) / [arm8](https://paddlelite-data.bj.bcebos.com/Release/2.8-rc/iOS/inference_lite_lib.ios.armv8.with_cv.with_extra.tiny_publish.tar.gz)|
**NOTE**:
@ -78,8 +78,9 @@ Paddle-Lite provides a variety of strategies to automatically optimize the origi
* Use pip to install Paddle-Lite. The following command uses `pip3.7` .
```shell
pip install paddlelite
pip install paddlelite==2.8
```
**Note**：The version of `paddlelite`'s wheel must match that of inference lib.
* Use `paddle_lite_opt` to optimize inference model, the parameters of `paddle_lite_opt` are as follows:
@ -121,17 +122,14 @@ Taking the `MobileNetV3_large_x1_0` model of PaddleClas as an example, we will i
```shell
# enter PaddleClas root directory
cd PaddleClas_root_path
export PYTHONPATH=$PWD
# download and uncompress the pre-trained model
wget https://paddle-imagenet-models-name.bj.bcebos.com/MobileNetV3_large_x1_0_pretrained.tar
tar -xf MobileNetV3_large_x1_0_pretrained.tar
# download and uncompress the inference model
wget https://paddle-imagenet-models-name.bj.bcebos.com/dygraph/inference/MobileNetV3_large_x1_0_infer.tar
tar -xf MobileNetV3_large_x1_0_infer.tar
# export the pre-trained model as an inference model
python tools/export_model.py -m MobileNetV3_large_x1_0 -p ./MobileNetV3_large_x1_0_pretrained/ -o ./MobileNetV3_large_x1_0_inference/
# convert inference model to Paddle-Lite optimized model
paddle_lite_opt --model_file=./MobileNetV3_large_x1_0_inference/model --param_file=./MobileNetV3_large_x1_0_inference/params --optimize_out=./MobileNetV3_large_x1_0
paddle_lite_opt --model_file=./MobileNetV3_large_x1_0_infer/inference.pdmodel --param_file=./MobileNetV3_large_x1_0_infer/inference.pdiparams --optimize_out=./MobileNetV3_large_x1_0
```
When the above code command is completed, there will be ``MobileNetV3_large_x1_0.nb` in the current directory, which is the converted model file.