Karina Zadorozhny
Karina Zadorozhny
@Manula99 yes, pytorch profiler or lightning profilers would be great!
@ncfrey has this example of how to generate synthetic data: https://github.com/prescient-design/lobster/blob/main/docs/RL_TRAINING.md#step-1-generate-synthetic-dataset And yes, a dockerfile would be great! @ncfrey and I will add contributions guidelines (I'm not sure if @Manula99...
MR to add contribution guidelines: #133
thank you!
looks great! could you pls add a usage example to `examples`?
`token_selection` should ideally be inside `src/lobster` or `src/lobster/model/utils` and the script in `cmdline`. Also, it would help if the script didn't implement any actual logic but just called functions defined...
could you pls add tests?
@kleinhenz is this ready to be merged? 🚀