Flight Log / spencerduncan/jimbot #362
feat: Implement dynamic observation space for Balatro RL training (#358) — a Claude Code (Anthropic) flight by spencerduncan
@spencerduncan opened this Claude Code (Anthropic) pull request in spencerduncan/jimbot on Jul 9, 2025; merged in 11.4 h.
Flown by
- Lines
- —
- Files
- —
- Commits
- —
- Reviews
- —
- To merge
- 11.4 h
From the pull request description
## Summary - Implemented DynamicObservationSpace class that handles variable-sized game states with attention masking - Supports extreme game states (100+ card hands) common in endless mode - Updated BalatroEnv to use the new dynamic observation space ## Changes - Created `DynamicObservationSpace` …
Excerpt (first 300 characters) — the full description is on GitHub.
View the pull request on GitHub →
This is your flight? Claim it — sign in with GitHub. Ownership is proven by the sign-in: the account that opened the PR is the only one that can claim it.
Read next
Cite this flight: AgentRiders Flight Log, spencerduncan/jimbot#362, record updated Nov 3, 2026, https://agentriders.com/logs/spencerduncan/jimbot/362/
Data from public GitHub activity; this page is not affiliated with GitHub or the repository.