Instructions to use adamlizerbram/cs654-final-project with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Laya
How to use adamlizerbram/cs654-final-project with Laya:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
CS654 Final Project: model weights
These are the final weights for a CS 654 (Reinforcement Learning) course project. The project compares DQN, Laya, and a Laya-shielded DQN on highway-env (highway-fast-v0).
| Folder | Contents |
|---|---|
laya-highway/ |
Laya, fine-tuned to judge driving actions as safe or unsafe and to choose actions |
Download
hf download adamlizerbram/cs654-final-project --include "laya-highway/*" --local-dir models/final # -> models/final/laya-highway
import laya
agent = laya.load("adamlizerbram/cs654-final-project", subfolder="laya-highway")
laya-highway
One checkpoint answers two questions about a text description of the highway around the ego car:
- Safety (
noul): will a proposed action lead to a collision or a gap under 5 m within 3 s? This yes/no answer drives the safety shield. - Action (
choice): which legal action is safest while still making good progress? This is the Laya-alone driving policy.
Data. Labels come from simulation, not humans. For each visited state and each legal action, a copy of the simulator applies the action and then keeps lane and speed for 2 more steps.
- An action is unsafe if the copy crashes or the same-lane gap drops below 5 m.
- The action label is the safe action with the most forward progress.
- Train episodes use seeds 0β299 and validation episodes use seeds 10000β10049. Evaluation seeds (100000+) were never used for training or tuning.
Training. laya-train from laya 0.4.30 on an A100:
- 4 epochs with the default
rlcdobjective,--shuffle-options, and an effective batch of 64 - 16,805 train items and 400 calibration items
- Temperature scaling fitted T = 5.0 for both question types
Validation results (before β after fine-tuning):
| Question | Items | Accuracy | ECE |
|---|---|---|---|
| Safety | 2,754 | 0.503 β 0.947 | 0.303 β 0.025 |
| Action | 697 | 0.144 β 0.891 | 0.586 β 0.048 |
The fine-tuned P(unsafe) is nearly binary: 99% of predictions fall below 0.1 or at 0.9 and above, and both clusters are well calibrated. 6.2% of truly unsafe actions are scored below 0.1.
Limitations.
- The labels imitate a 3-second lookahead rule, not an optimal driver.
highway-fast-v0uses coarse 5 Hz physics.- The model reads a fixed text format, so other state descriptions are out of distribution.
- Downloads last month
- -
Model tree for adamlizerbram/cs654-final-project
Base model
convaiinnovations/laya