Update README.md
Browse files
README.md
CHANGED
@@ -25,6 +25,18 @@ model-index:
|
|
25 |
This is a trained model of a **PPO** agent playing **LunarLander-v2**
|
26 |
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
|
27 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
28 |
## Usage (with Stable-baselines3)
|
29 |
TODO: Add your code
|
30 |
|
|
|
25 |
This is a trained model of a **PPO** agent playing **LunarLander-v2**
|
26 |
using the [stable-baselines3 library](https://github.com/DLR-RM/stable-baselines3).
|
27 |
|
28 |
+
Made as part of the Deep RL course: https://huggingface.co/learn/deep-rl-course. Tuned with Optuna, as introduced in the course. This is my first successful attempt of using Optuna, so do not expect the code or parameters to be ideal!
|
29 |
+
|
30 |
+
I was able to improve upon my result from Unit1, https://huggingface.co/humnrdble/DeepRL-unit1. Both models were trained for 1500000 steps. The video of my first attempt certainly looks smoother, but scores worse.
|
31 |
+
|
32 |
+
The code is available in unit1-notebook-tuned.ipynb, but no attempt was made to make it particularly legible.
|
33 |
+
|
34 |
+
Hyperparameters deviating from the Stable-baselines3 baseline:
|
35 |
+
- gamma: 1-0.006075594024321983
|
36 |
+
- max_grad_norm: 1.8559426752164974
|
37 |
+
- exponent_n_steps: 9 (i.e. 2**9 steps)
|
38 |
+
- learning_rate: 0.0011176199638550707
|
39 |
+
|
40 |
## Usage (with Stable-baselines3)
|
41 |
TODO: Add your code
|
42 |
|