Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -22,15 +22,15 @@ PPO-trained agent for [OpenFront.io](https://openfront.io), a multiplayer territ
|
|
| 22 |
- **Parallel envs:** 8
|
| 23 |
- **Learning rate:** 0.00015
|
| 24 |
- **Rollout steps:** 512
|
| 25 |
-
- **Updates trained:**
|
| 26 |
-
- **Global steps:**
|
| 27 |
-
- **Best mean reward:**
|
| 28 |
|
| 29 |
## Final Training Metrics
|
| 30 |
|
| 31 |
-
- **Mean reward:**
|
| 32 |
-
- **Mean episode length:**
|
| 33 |
-
- **Loss:**
|
| 34 |
|
| 35 |
## Usage
|
| 36 |
|
|
|
|
| 22 |
- **Parallel envs:** 8
|
| 23 |
- **Learning rate:** 0.00015
|
| 24 |
- **Rollout steps:** 512
|
| 25 |
+
- **Updates trained:** 2750
|
| 26 |
+
- **Global steps:** 11264000
|
| 27 |
+
- **Best mean reward:** 2761.6309762120245
|
| 28 |
|
| 29 |
## Final Training Metrics
|
| 30 |
|
| 31 |
+
- **Mean reward:** 1866.676198823452
|
| 32 |
+
- **Mean episode length:** 35228.65
|
| 33 |
+
- **Loss:** 17.78866958618164
|
| 34 |
|
| 35 |
## Usage
|
| 36 |
|