How It Works
Structure
This is where the repo becomes a personal notebook: loose utilities, cloned variants, abandoned attempts, policy files, clutter folders, and partially broken experiments.
Rough experiments + dead ends + utility scraps
Archive and lab-notebook material
Useful for your personal review, but not something to treat as production-grade code without curation.
Archive first
Unreliable as-is
Why Reuse
Role in the new DRL arm
It still matters because it captures your own reasoning trail and the branches you explored after the coursework. That is valuable review material, even when the code is not clean enough to power the live app directly.
Key Assets
What is already here
Lunar Branch
experiment branchDQN, DDPG, PPO, and hybrid experiments around LunarLander.
source-material/lunar
Pong Policies
model artifactsSaved `.policy` checkpoints from old Pong runs.
source-material/reinforce/pong-PPO-REINFORCE/data
Utility Scratchpad
python moduleSmall snippets and helper fragments collected outside the main folders.
source-material/drl_utils.py
DDPG Clutter
experiment branchOld or abandoned Reacher variants that still record your exploration path.
source-material/classwork/project-reports/p2_continuous-control/ddpg/clutter