REnforce
REnforce copied to clipboard
NivenT
Reinforcement learning library written in Rust
As a bare minimum for thinking a new RL algorithm was possible implemented correctly, it is given a [test](https://github.com/NivenT/REnforce/blob/master/tests/bandit.rs) on the [N-armed bandit](https://www.wikiwand.com/en/Multi-armed_bandit) problem. This environment is about as simple...
I guess first off, ideally there would be an [example](https://github.com/NivenT/REnforce/tree/master/examples) for every training algorithm added. For now, the best way I've found to set these up seems to be to...
How I currently envision this working is by making a custom error type, and then having most user-facing functions return `Result`s. I like how [rusty-machine](https://github.com/AtheMathmo/rusty-machine) has their [error type](https://github.com/AtheMathmo/rusty-machine/blob/master/src/learning/error.rs#L13) set...
Basically what the title says; there needs to be more documenting of the code. Everything is required to have some documentation or the compiler will yell at you, but the...
The cartpole example works fine when I run the debug build, but if I run with `--release`, there seems to be a communication problem with the gym server: ``` $...
