added inverted pendulum and inverted double pendulum

mihirp98 · Apr 7, 2020 · 16e9302 · 16e9302
1 parent 7ffb5e4
commit 16e9302
Show file tree

Hide file tree

Showing 4 changed files with 110 additions and 0 deletions.
diff --git a/examples/config/inverted_double_pendulum/0.py b/examples/config/inverted_double_pendulum/0.py
@@ -0,0 +1,37 @@
+params = {
+    'type': 'MBPO',
+    'universe': 'gym',
+    'domain': 'InvertedDoublePendulum',
+    'task': 'v2',
+
+    'log_dir': '~/ray_mbpo/',
+    'exp_name': 'defaults',
+
+    'kwargs': {
+        'n_epochs': 80, ## 20k steps
+        'epoch_length': 250,
+        'train_every_n_steps': 1,
+        'n_train_repeat': 20,
+        'eval_render_mode': None,
+        'eval_n_episodes': 1,
+        'eval_deterministic': True,
+
+        'discount': 0.99,
+        'tau': 5e-3,
+        'reward_scale': 1.0,
+
+        'model_train_freq': 250,
+        'model_retain_epochs': 1,
+        'rollout_batch_size': 100e3,
+        'deterministic': False,
+        'num_networks': 7,
+        'num_elites': 5,
+        'real_ratio': 0.05,
+        'target_entropy': -0.5,
+        'max_model_t': None,
+        'rollout_schedule': [1, 15, 1, 1],
+        'hidden_dim': 200,
+        'n_initial_exploration_steps': 500,
+    }
+}
+
diff --git a/examples/config/inverted_pendulum/0.py b/examples/config/inverted_pendulum/0.py
@@ -0,0 +1,36 @@
+params = {
+    'type': 'MBPO',
+    'universe': 'gym',
+    'domain': 'InvertedPendulum',
+    'task': 'v2',
+
+    'log_dir': '~/ray_mbpo/',
+    'exp_name': 'defaults',
+
+    'kwargs': {
+        'n_epochs': 80, ## 20k steps
+        'epoch_length': 250,
+        'train_every_n_steps': 1,
+        'n_train_repeat': 10,
+        'eval_render_mode': None,
+        'eval_n_episodes': 1,
+        'eval_deterministic': True,
+
+        'discount': 0.99,
+        'tau': 5e-3,
+        'reward_scale': 1.0,
+
+        'model_train_freq': 250,
+        'model_retain_epochs': 1,
+        'rollout_batch_size': 100e3,
+        'deterministic': False,
+        'num_networks': 7,
+        'num_elites': 5,
+        'real_ratio': 0.05,
+        'target_entropy': -0.05,
+        'max_model_t': None,
+        'rollout_schedule': [1, 15, 1, 1],
+        'hidden_dim': 200,
+        'n_initial_exploration_steps': 500,
+    }
+}
diff --git a/mbpo/static/inverted_double_pendulum.py b/mbpo/static/inverted_double_pendulum.py
@@ -0,0 +1,20 @@
+import sys
+import numpy as np
+import pdb
+
+class StaticFns:
+
+    @staticmethod
+    def termination_fn(obs, act, next_obs):
+        assert len(obs.shape) == len(next_obs.shape) == len(act.shape) == 2
+
+        sin1, cos1 = next_obs[:,1], next_obs[:,3]
+        sin2, cos2 = next_obs[:,2], next_obs[:,4]
+        theta_1 = np.arctan2(sin1, cos1)
+        theta_2 = np.arctan2(sin2, cos2)
+        y = 0.6 * (cos1 + np.cos(theta_1 + theta_2))
+
+        done = y <= 1
+
+        done = done[:,None]
+        return done
diff --git a/mbpo/static/inverted_pendulum.py b/mbpo/static/inverted_pendulum.py
@@ -0,0 +1,17 @@
+import sys
+import numpy as np
+import pdb
+
+class StaticFns:
+
+    @staticmethod
+    def termination_fn(obs, act, next_obs):
+        assert len(obs.shape) == len(next_obs.shape) == len(act.shape) == 2
+
+        notdone = np.isfinite(next_obs).all(axis=-1) \
+        		  * (np.abs(next_obs[:,1]) <= .2)
+        done = ~notdone
+
+        done = done[:,None]
+
+        return done