Reinforcement Learning
sample-factory
deep-reinforcement-learning
vizdoom
doom_health_gathering_supreme
Eval Results (legacy)
Instructions to use johith9381/rl_course_vizdoom_health_gathering_supreme with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sample-factory
How to use johith9381/rl_course_vizdoom_health_gathering_supreme with sample-factory:
python -m sample_factory.huggingface.load_from_hub -r johith9381/rl_course_vizdoom_health_gathering_supreme -d ./train_dir
- Notebooks
- Google Colab
- Kaggle
VizDoom Health Gathering Supreme
Sample Factory APPO agent trained for the doom_health_gathering_supreme environment.
Certification
| Metric | Result |
|---|---|
| Environment | doom_health_gathering_supreme |
| Required minimum true objective | 5.0 |
| Evaluated true objective | 5.609 |
| Status | PASS |
Model
- Algorithm: APPO
- Train step: 124
- Environment steps: 507904
- Checkpoint:
checkpoint_000000124_507904.pth
Evaluation
The trained policy was evaluated with deterministic action selection.
True objective: 5.609
This model exceeds the required certification threshold of 5.0.
- Downloads last month
- -
Evaluation results
- mean_reward on doom_health_gathering_supremeself-reported5.609