Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,38 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- reinforcement-learning
|
| 5 |
+
- stable-baselines3
|
| 6 |
+
- mujoco
|
| 7 |
+
- ant-v4
|
| 8 |
+
- ppo
|
| 9 |
+
pipeline_tag: reinforcement-learning
|
| 10 |
+
library_name: stable-baselines3
|
| 11 |
+
model_name: PPO-Ant-v4
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# PPO - Ant-v4 🌟
|
| 15 |
+
|
| 16 |
+
A Proximal Policy Optimization (PPO) agent trained with **stable-baselines3** on the MuJoCo **`Ant-v4`** environment.
|
| 17 |
+
|
| 18 |
+
| | Details |
|
| 19 |
+
|---|---|
|
| 20 |
+
| Environment | `gymnasium==0.29` & `mujoco==2.3` (`Ant-v4`) |
|
| 21 |
+
| Algorithm | PPO (`stable-baselines3==2.3.0`) |
|
| 22 |
+
| Timesteps | **100 000** |
|
| 23 |
+
| Policy | `MlpPolicy` *(2 × 64 hidden, tanh)* |
|
| 24 |
+
| Return (mean ± std) | ~ *964* |
|
| 25 |
+
| Seed | `0` |
|
| 26 |
+
|
| 27 |
+
## Hyper-parameters
|
| 28 |
+
|
| 29 |
+
```jsonc
|
| 30 |
+
{
|
| 31 |
+
"n_steps": 128,
|
| 32 |
+
"batch_size": 64,
|
| 33 |
+
"n_epochs": 20,
|
| 34 |
+
"gamma": 0.99,
|
| 35 |
+
"learning_rate": 3e-4,
|
| 36 |
+
"ent_coef": 0.0,
|
| 37 |
+
"clip_range": 0.2
|
| 38 |
+
}
|