Closed Luka-Loncar closed 2 months ago
@Luka-Loncar I suspect I'll own this unless @teslashibe says otherwise. Also, I changed the ticket this morning reflect we'll try and keep experimentation notebooks in Deepnote
Great to hear! If you want to pick it up that is more than welcome.
Blocked by https://github.com/masa-finance/masa-bittensor/issues/65
To clarify, this ticket is issue 65, yeah? Can you point me to the blocker or is it resolved?
@Luka-Loncar The project has been created here. I just want to confirm the repo we want it to be associated with before finalizing the link. masa-bittensor is public.
Good question, I would ask @mudler to provide suggestions on this. We are usually using private repos if we don't want something to be visible.
Blocked by #65
To clarify, this ticket is issue 65, yeah? Can you point me to the blocker or is it resolved?
sorry - it was meant to be #11 . The final simulation would have to take into account the result of the spike (as we are trying to figure out a reward mechanism, and this is a simulation of it).
@Luka-Loncar The project has been created here. I just want to confirm the repo we want it to be associated with before finalizing the link. masa-bittensor is public.
I'm not sure there is anything we need to keep private, as the work is already happening in the open, so I'd say it should belong to this repository - however I'm ok to keep it private for now and use the roadmap repository for this until we finalize the simulation and is ready to be shared.
Ok, lets do it like that.
@theMultitude is there any triage or any other info that you require from product in order to tackle this one?
@mudler in my mind this information isn't necessary as part of this public repo and would sit more neatly in a data science/experimentation repo that won't need to be surfaced in the same way. It also helps me to orchestrate and support experimentation in a more centralized way if we have a dedicated repo for it. Experiments and methodology can be surfaced through publications once a bit more polished.
@mudler in my mind this information isn't necessary as part of this public repo and would sit more neatly in a data science/experimentation repo that won't need to be surfaced in the same way. It also helps me to orchestrate and support experimentation in a more centralized way if we have a dedicated repo for it. Experiments and methodology can be surfaced through publications once a bit more polished.
makes sense here :+1:
@Luka-Loncar a pull request to add the notebook. https://github.com/masa-finance/masa-bittensor/pull/114
I think we may want to centralize this type of work sooner rather than later as mentioned above but I didn't want to block y'all from getting started.
Thanks!
What would that centralization look like? Can you please explain it in a bit more detail?
@Luka-Loncar As mentioned here I'd like to have a repo dedicated to data science work at Masa. It would make it easier for me to consult and contribute to related types of work.
I don't think that it would be any kind of issue that you have your own repo.
@mudler do you have anything against that maybe?
@theMultitude can we close this ticket now?
@theMultitude sounds good here - can you create the repo? or shall I?
@mudler I can tomorrow morning. @Luka-Loncar I'll close it when I finish.
I see also https://github.com/masa-finance/masa-bittensor/pull/114 still open tied to this
I think since the #114 PR is merged, we can close this
After we have completed our spike in https://github.com/masa-finance/masa-bittensor/issues/11, we want to have a simulation of the rewards in a notebook so we can visualize the results and sharing those around (with the community, and other stakeholders).
Action Items