History

石晓伟 d008260fa8 update the initialization of anakin subgraph (#17880 ) test=develop		6 years ago
..
demo_ci	fix the random compilation failure on windows test=develop (#17475 )	6 years ago
details	…
CMakeLists.txt	Capi for a ngraph engine (#17037 )	6 years ago
README.md	…
analysis_config.cc	update the initialization of anakin subgraph (#17880 )	6 years ago
analysis_predictor.cc	fix: when use the load model from memory mode, the RAM occupy is high (#17788 )	6 years ago
analysis_predictor.h	…
analysis_predictor_tester.cc	INT8 MKL-DNN v2 integrate to slim (#17634 )	6 years ago
api.cc	…
api_anakin_engine.cc	…
api_anakin_engine.h	…
api_impl.cc	…
api_impl.h	…
api_impl_tester.cc	…
api_tester.cc	…
helper.cc	…
helper.h	fix security bugs : (#17464 )	6 years ago
high_level_api.md	…
high_level_api_cn.md	…
mkldnn_quantizer.cc	Improve mobilenetv2 INT8 performance by using INT8 relu as post-op (#17570 )	6 years ago
mkldnn_quantizer.h	INT8 MKL-DNN v2 integrate to slim (#17634 )	6 years ago
mkldnn_quantizer_config.cc	add Concat quantization (#17448 )	6 years ago
paddle_anakin_config.h	…
paddle_analysis_config.h	fix: when use the load model from memory mode, the RAM occupy is high (#17788 )	6 years ago
paddle_api.h	…
paddle_inference_api.h	…
paddle_mkldnn_quantizer_config.h	…
paddle_pass_builder.cc	fix: when use the load model from memory mode, the RAM occupy is high (#17788 )	6 years ago
paddle_pass_builder.h	Capi for a ngraph engine (#17037 )	6 years ago

README.md

Embed Paddle Inference in Your Application

Paddle inference offers the APIs in C and C++ languages.

You can easily deploy a model trained by Paddle following the steps as below:

Optimize the native model;
Write some codes for deployment.

The APIs

All the released APIs are located in the paddle_inference_api.h header file. The stable APIs are wrapped by namespace paddle, the unstable APIs are protected by namespace paddle::contrib.

Write some codes

Read paddle_inference_api.h for more information.