Harshit Sharma

Is anyone working on this? if not i can raise a pr, want to start contributing to this project !

I see there's already a partial `SQLiteDBClient` implementation in `db_client.py` (about 6 methods need SQLite conversion). I'll complete this implementation to make SQLite fully functional, then we can decide on...

okay I see it, my bad! I hadnt fully gone through the readme and was just going off based on the code itself. Let me run everything end to end,...

I wanted to think of the actions from the perspective of mutations on the game state. #### Game State (`S_T`) At any tick `T`, the full game state can be...

imo keeping domain explicit is a structural prior that: 1. Reduces effective action entropy, if the true distribution has multiple-modes (map ops vs. inventory ops vs. research ops), modelling it...

> Not to state the obvious, but I would do what I suggested a few weeks back. 😅 > > In this context I don't see the benefit of splitting...

for FLE: the following observation/helpers probably can be implemented easily: - `get_research_progress` - `inspect_inventory` - `score` - `get_prototype_recipe` actions that should be replaced by some broader combination of vision +...

then render + a combined querying method should be good enough for traces to also record observations. but the replay and the headless renders would need to possibly look the...

ahh yes ive seen the render. but im saying that for traces from replays we have immediate access to screenshots from the game directly already, so we may want to...

that works, i just wished we could use the client for this as the existing rendering pipeline is so much more reliable. there is a POC here to convert the...