shivakanthsujit
commited on
Commit
·
dad54c9
1
Parent(s):
1260824
Init Commit
Browse files- README.md +19 -0
- config.json +20 -0
- env_stats.pickle +3 -0
- model.pth +3 -0
README.md
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: mbrl-lib
|
3 |
+
tags:
|
4 |
+
- mbrl-continuous-cartpole
|
5 |
+
- deep-reinforcement-learning
|
6 |
+
- reinforcement-learning
|
7 |
+
- mbrl-lib
|
8 |
+
---
|
9 |
+
|
10 |
+
# **OneDTransitionRewardModel** Agent playing **mbrl-continuous-cartpole**
|
11 |
+
This is a trained model of a **OneDTransitionRewardModel** agent playing **mbrl-continuous-cartpole**
|
12 |
+
using [MBRL-Lib](https://github.com/facebookresearch/mbrl-lib).
|
13 |
+
|
14 |
+
## Usage (with MBRL-Lib)
|
15 |
+
TODO: Add your code
|
16 |
+
```python
|
17 |
+
from mbrl import ...
|
18 |
+
...
|
19 |
+
```
|
config.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
dynamics_model:
|
2 |
+
_target_: mbrl.models.GaussianMLP
|
3 |
+
num_layers: 3
|
4 |
+
ensemble_size: 5
|
5 |
+
device: cpu
|
6 |
+
hid_size: 200
|
7 |
+
in_size: 5
|
8 |
+
out_size: 4
|
9 |
+
deterministic: false
|
10 |
+
propagation_method: fixed_model
|
11 |
+
activation_fn_cfg:
|
12 |
+
_target_: torch.nn.LeakyReLU
|
13 |
+
negative_slope: 0.01
|
14 |
+
algorithm:
|
15 |
+
learned_rewards: false
|
16 |
+
target_is_delta: true
|
17 |
+
normalize: true
|
18 |
+
overrides:
|
19 |
+
model_batch_size: 32
|
20 |
+
validation_ratio: 0.05
|
env_stats.pickle
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:69554307f755e9042cf296a7c89d744651bea8ae81d226f8e3150b50f8e8ac01
|
3 |
+
size 238
|
model.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ad047315fd225bd9999164bf547ea17ddf7d2f976f4994dc2f4fdfed69dc309f
|
3 |
+
size 1667439
|