- A next-generation dynamic and high-performance language for AI and IOT with natural born distributed computing ability
- A super glue to easily integrating with other languages such as c++/c, python and javascript and any framework cross operation system barriers.
- Running faster than python about 3x-5x
- fully optimized tensor computing architecture
- easily build neural network with tensor expression
- automatically generate tensor data flow graph and compile for target
- boost inference/training performance about 6x-10x in GPU(CUDA)
- build from Windows
- git clone https://github.com/xlang-foundation/xlang.git
- use Visual Studio to open this xlang folder
- choose configuration for example Local Machine/x64-Debug, WSL:Ubuntu/WSL-GCC-Debug
- build ( click on Visual Studio's menu: build/build all)
- build from Linux(Ubuntu)
sudo apt-get install uuid-dev
sudo apt-get install libssl-dev
sudo apt-get install python3-dev and also need to pip install numpy if not want to enable this feature, just comment out line below in CMakeLists.txt in root folder add_subdirectory("PyEng")
1. git clone https://github.com/xlang-foundation/xlang.git 2. cd xlang 3. mkdir out 4. cd out 5. cmake .. 6. make
- go to console window, cd to xlang executable file folder
- xlang
- install xlang plugin in vscode
- start xlang with parameter: -event_loop -dbg -enable_python
xlang -event_loop -dbg -enable_python - open or new a file with .x extension name
click on vs code menu run/start debuging then vscode will automatcilly connect with xlang to run
- in Windows, install Android Studio
- open project from folder xlang\Android
- then build from Android Stduio menu: Build/Make Project