TastyPancakes/Flux.jl
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

and again

Browse Source
This commit is contained in:
Mike J Innes 2016-05-08 20:05:51 +01:00
parent 3bbc660093
commit b5b70ba4cc
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
README.md
View File

@ -4,6 +4,7 @@ Flux is an experimental machine perception / ANN library for Julia. It's most si
Flux has a few key differences from other libraries:
* Flux's [graph-based DSL](https://github.com/MikeInnes/Flow.jl), which provides optimisations and automatic differentiation (in the spirit of Theano), is very tightly integrated with the language. This means nice syntax for your equations (`σ(W*x+b)` anyone?) and no unwieldy `compile` steps.
* The graph DSL directly represents models, as opposed to computations, so custom architectures – and in particular, recurrent architectures – are expressed very naturally.
* Flux is written in [Julia](http://julialang.org), which means there's no "dropping down" to C – it's Julia all the way down, and you can prototype both high-level architectures and high-performance GPU kernels from the same language.
* Flux's [graph-based DSL](https://github.com/MikeInnes/Flow.jl), which provides optimisations and automatic differentiation (à la Theano), is very tightly integrated with the language. This means nice syntax for your equations (`σ(W*x+b)` anyone?) and no unwieldy `compile` steps.
* The graph DSL directly represents models, as opposed to computations, so custom architectures – and in particular, recurrent models – are easy to express.
* Those fancy features are completely optional. You can implement arbitrary functionality in a Torch-like fashion if you wish, since layers are simply objects that satisfy a small interface (à la Torch).
* Flux is written in [Julia](http://julialang.org), which means there's no "dropping down" to C. It's Julia all the way down, and you can prototype both high-level architectures and high-performance GPU kernels from the same language. This also makes the library itself very easy to understand and extend.