git
ask
hub
Privacy
Terms
Sign in with GitHub
qianzhong-chen/openspiral
↗
Stars ·
21
Language ·
Python
License ·
Apache-2.0
Simplify and Visualize This Repo
Scan the safety of this repo
Self Host this repo
Check out more work of Developer: qianzhong-chen
Ask anything about this repo to start.
I have a behavior-cloned pi-zero policy and a SARM2-labeled LeRobot dataset. Walk me through running the SPIRAL residual RL training loop with train_residual_rl.py.
How does the hybrid critic target in SPIRAL blend TD3 bootstrapped value learning with Monte Carlo returns for long-horizon tasks?
What GPU memory is required to run SPIRAL residual RL training versus full pi-zero fine-tuning, and which GPUs are supported?
Full explanation on explaingit →
📎
Send
By chatting or signing in you agree to the
Terms
and chat-message logging (revocable in
History
).