A toy domain specific language for deep learning tensor expressions.
$ git clone git@github.com:lijiansong/Phaeton.git # or git clone https://github.com/lijiansong/Phaeton.git
$ cd Phaeton
$ mkdir build && cd build
$ cmake -DCMAKE_CXX_COMPILER=$(which g++) -DCMAKE_C_COMPILER=$(which gcc) .. && make -j
- taco: https://github.com/tensor-compiler/taco
- lift: https://github.com/lift-project/lift and its successor rise: https://github.com/rise-lang
- mlir: https://github.com/tensorflow/mlir
- halide: https://github.com/halide/Halide
- futhark: https://github.com/diku-dk/futhark
- tc: https://github.com/facebookresearch/TensorComprehensions
- glow: https://github.com/pytorch/glow
- tvm: https://github.com/dmlc/tvm
Note: Partial design concept of Phaeton comes from clang and mlir.
I only contribute to this project in my part-time, so this repo will be updated irregularly.
If you are interested in becoming a contributor, or you have any questions or suggestions, feel free to open an issue or drop me a message.