Hugging Face's logo

Datasets:
hc99
/
FEA-Bench

Dataset card Files
xet
 Community
FEA-Bench
File size: 4,504 Bytes
a89d35f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
.. _td3:

.. automodule:: stable_baselines3.td3


TD3
===

`Twin Delayed DDPG (TD3) <https://spinningup.openai.com/en/latest/algorithms/td3.html>`_ Addressing Function Approximation Error in Actor-Critic Methods.

TD3 is a direct successor of :ref:`DDPG <ddpg>` and improves it using three major tricks: clipped double Q-Learning, delayed policy update and target policy smoothing.
We recommend reading `OpenAI Spinning guide on TD3 <https://spinningup.openai.com/en/latest/algorithms/td3.html>`_ to learn more about those.


.. rubric:: Available Policies

.. autosummary::
    :nosignatures:

    MlpPolicy
    CnnPolicy


Notes
-----

- Original paper: https://arxiv.org/pdf/1802.09477.pdf
- OpenAI Spinning Guide for TD3: https://spinningup.openai.com/en/latest/algorithms/td3.html
- Original Implementation: https://github.com/sfujim/TD3

.. note::

    The default policies for TD3 differ a bit from others MlpPolicy: it uses ReLU instead of tanh activation,
    to match the original paper


Can I use?
----------

-  Recurrent policies: ❌
-  Multi processing: ❌
-  Gym spaces:


============= ====== ===========
Space         Action Observation
============= ====== ===========
Discrete      ❌      ✔️
Box           ✔️       ✔️
MultiDiscrete ❌      ✔️
MultiBinary   ❌      ✔️
============= ====== ===========


Example
-------

.. code-block:: python

  import gym
  import numpy as np

  from stable_baselines3 import TD3
  from stable_baselines3.td3.policies import MlpPolicy
  from stable_baselines3.common.noise import NormalActionNoise, OrnsteinUhlenbeckActionNoise

  env = gym.make('Pendulum-v0')

  # The noise objects for TD3
  n_actions = env.action_space.shape[-1]
  action_noise = NormalActionNoise(mean=np.zeros(n_actions), sigma=0.1 * np.ones(n_actions))

  model = TD3(MlpPolicy, env, action_noise=action_noise, verbose=1)
  model.learn(total_timesteps=10000, log_interval=10)
  model.save("td3_pendulum")
  env = model.get_env()

  del model # remove to demonstrate saving and loading

  model = TD3.load("td3_pendulum")

  obs = env.reset()
  while True:
      action, _states = model.predict(obs)
      obs, rewards, dones, info = env.step(action)
      env.render()

Results
-------

PyBullet Environments
^^^^^^^^^^^^^^^^^^^^^

Results on the PyBullet benchmark (1M steps) using 3 seeds.
The complete learning curves are available in the `associated issue #48 <https://github.com/DLR-RM/stable-baselines3/issues/48>`_.


.. note::

  Hyperparameters from the `gSDE paper <https://arxiv.org/abs/2005.05719>`_ were used (as they are tuned for PyBullet envs).


*Gaussian* means that the unstructured Gaussian noise is used for exploration,
*gSDE* (generalized State-Dependent Exploration) is used otherwise.

+--------------+--------------+--------------+--------------+
| Environments | SAC          | SAC          | TD3          |
+==============+==============+==============+==============+
|              | Gaussian     | gSDE         | Gaussian     |
+--------------+--------------+--------------+--------------+
| HalfCheetah  | 2757 +/- 53  | 2984 +/- 202 | 2774 +/- 35  |
+--------------+--------------+--------------+--------------+
| Ant          | 3146 +/- 35  | 3102 +/- 37  | 3305 +/- 43  |
+--------------+--------------+--------------+--------------+
| Hopper       | 2422 +/- 168 | 2262 +/- 1   | 2429 +/- 126 |
+--------------+--------------+--------------+--------------+
| Walker2D     | 2184 +/- 54  | 2136 +/- 67  | 2063 +/- 185 |
+--------------+--------------+--------------+--------------+


How to replicate the results?
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Clone the `rl-zoo repo <https://github.com/DLR-RM/rl-baselines3-zoo>`_:

.. code-block:: bash

  git clone https://github.com/DLR-RM/rl-baselines3-zoo
  cd rl-baselines3-zoo/


Run the benchmark (replace ``$ENV_ID`` by the envs mentioned above):

.. code-block:: bash

  python train.py --algo td3 --env $ENV_ID --eval-episodes 10 --eval-freq 10000


Plot the results:

.. code-block:: bash

  python scripts/all_plots.py -a td3 -e HalfCheetah Ant Hopper Walker2D -f logs/ -o logs/td3_results
  python scripts/plot_from_file.py -i logs/td3_results.pkl -latex -l TD3


Parameters
----------

.. autoclass:: TD3
  :members:
  :inherited-members:

.. _td3_policies:

TD3 Policies
-------------

.. autoclass:: MlpPolicy
  :members:
  :inherited-members:

.. autoclass:: stable_baselines3.td3.policies.TD3Policy
  :members:
  :noindex:

.. autoclass:: CnnPolicy
  :members: