Reducing compiling time of forward/backward by making a global cache of functions

hikettei / cl-waffe2

[Experimental] Graph and Tensor Abstraction for Deep Learning all in Common Lisp

https://hikettei.github.io/cl-waffe2/

MIT License

133 stars 5 forks source link

Closed hikettei closed 1 year ago

hikettei commented 1 year ago

Issue #4 This Pull Request satisfies my requirement < 5sec, but I can do much more...