Spaces:
Build error
Build error
Update app.py
Browse files
app.py
CHANGED
@@ -7,9 +7,9 @@ from huggingface_sb3 import load_from_hub
|
|
7 |
from stable_baselines3 import PPO
|
8 |
from stable_baselines3.common.env_util import make_atari_env
|
9 |
from stable_baselines3.common.vec_env import VecFrameStack
|
10 |
-
|
11 |
from stable_baselines3.common.env_util import make_atari_env
|
12 |
|
|
|
13 |
st.title("Atari Environments Live Model")
|
14 |
|
15 |
# @st.cache This is not cachable :(
|
@@ -36,10 +36,10 @@ def load_model(env_name):
|
|
36 |
|
37 |
return model
|
38 |
|
39 |
-
st.
|
40 |
|
41 |
-
st.
|
42 |
-
st.
|
43 |
|
44 |
env_name = st.selectbox(
|
45 |
"Select environment",
|
|
|
7 |
from stable_baselines3 import PPO
|
8 |
from stable_baselines3.common.env_util import make_atari_env
|
9 |
from stable_baselines3.common.vec_env import VecFrameStack
|
|
|
10 |
from stable_baselines3.common.env_util import make_atari_env
|
11 |
|
12 |
+
|
13 |
st.title("Atari Environments Live Model")
|
14 |
|
15 |
# @st.cache This is not cachable :(
|
|
|
36 |
|
37 |
return model
|
38 |
|
39 |
+
st.write("In game theory and optimization Nash Equilibrium loss minimization starts playing randomly but then by understanding ratios of action success to action-reward with an action (observe, decide/predict, act and then observe outcome the Deep RL agents go from 50% efficiency to 98-99% efficiency based on quality of decision without making mistakes.")
|
40 |
|
41 |
+
st.write("list of agent environments https://github.com/DLR-RM/rl-baselines3-zoo/blob/master/benchmark.md")
|
42 |
+
st.write("Deep RL models: https://huggingface.co/sb3")
|
43 |
|
44 |
env_name = st.selectbox(
|
45 |
"Select environment",
|