model

Browse files

Files changed (11) hide show

README.md +1 -1
config.json +0 -0
humanoid-v5-sac-medium.zip +2 -2
humanoid-v5-sac-medium/actor.optimizer.pth +1 -1
humanoid-v5-sac-medium/critic.optimizer.pth +1 -1
humanoid-v5-sac-medium/data +25 -25
humanoid-v5-sac-medium/ent_coef_optimizer.pth +1 -1
humanoid-v5-sac-medium/policy.pth +1 -1
humanoid-v5-sac-medium/pytorch_variables.pth +1 -1
replay.mp4 +2 -2
results.json +1 -1

README.md CHANGED Viewed

@@ -16,7 +16,7 @@ model-index:
       type: Humanoid-v5
     metrics:
     - type: mean_reward
-      value: 4731.38 +/- 1469.86
       name: mean_reward
       verified: false
 ---

       type: Humanoid-v5
     metrics:
     - type: mean_reward
+      value: 6846.05 +/- 18.34
       name: mean_reward
       verified: false
 ---

config.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

humanoid-v5-sac-medium.zip CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b33e3ad0b19a5287f624e3c9b645b2c76be5a5dbcf6ab7517f9f90b73be6bd1d
-size 7179831

 version https://git-lfs.github.com/spec/v1
+oid sha256:3b92da2a5bd8b4e7024f7f971ac5ef2fdcfc49caf75e3edc3b826d216317c0c5
+size 7179839

humanoid-v5-sac-medium/actor.optimizer.pth CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d7b4d49e5d237d57e2f6f5ce1746690d1355890c4ede08d183d0a643593c9969
 size 1317966

 version https://git-lfs.github.com/spec/v1
+oid sha256:2ff3dc54ca642f8fa9030366ea4764eb6b2ed5a4a7ae37e1910fdce752aed867
 size 1317966

humanoid-v5-sac-medium/critic.optimizer.pth CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9dc7a7ae944adf9b0b8896ce55476c377646cbcf06bd94e54ab658c438b6c785
 size 2566058

 version https://git-lfs.github.com/spec/v1
+oid sha256:8a5659fbac9aa4693598f5321798baa8866618f4daa79b3f9f2129dc883fa7f1
 size 2566058

humanoid-v5-sac-medium/data CHANGED Viewed

@@ -5,33 +5,33 @@
         "__module__": "stable_baselines3.sac.policies",
         "__annotations__": "{'actor': <class 'stable_baselines3.sac.policies.Actor'>, 'critic': <class 'stable_baselines3.common.policies.ContinuousCritic'>, 'critic_target': <class 'stable_baselines3.common.policies.ContinuousCritic'>}",
         "__doc__": "\n    Policy class (with both actor and critic) for SAC.\n\n    :param observation_space: Observation space\n    :param action_space: Action space\n    :param lr_schedule: Learning rate schedule (could be constant)\n    :param net_arch: The specification of the policy and value networks.\n    :param activation_fn: Activation function\n    :param use_sde: Whether to use State Dependent Exploration or not\n    :param log_std_init: Initial value for the log standard deviation\n    :param use_expln: Use ``expln()`` function instead of ``exp()`` when using gSDE to ensure\n        a positive standard deviation (cf paper). It allows to keep variance\n        above zero and prevent it from growing too fast. In practice, ``exp()`` is usually enough.\n    :param clip_mean: Clip the mean output when using gSDE to avoid numerical instability.\n    :param features_extractor_class: Features extractor to use.\n    :param features_extractor_kwargs: Keyword arguments\n        to pass to the features extractor.\n    :param normalize_images: Whether to normalize images or not,\n         dividing by 255.0 (True by default)\n    :param optimizer_class: The optimizer to use,\n        ``th.optim.Adam`` by default\n    :param optimizer_kwargs: Additional keyword arguments,\n        excluding the learning rate, to pass to the optimizer\n    :param n_critics: Number of critic networks to create.\n    :param share_features_extractor: Whether to share or not the features extractor\n        between the actor and the critic (this saves computation time)\n    ",
-        "__init__": "<function SACPolicy.__init__ at 0x7f984c210400>",
-        "_build": "<function SACPolicy._build at 0x7f984c210a40>",
-        "_get_constructor_parameters": "<function SACPolicy._get_constructor_parameters at 0x7f984c210ae0>",
-        "reset_noise": "<function SACPolicy.reset_noise at 0x7f984c210b80>",
-        "make_actor": "<function SACPolicy.make_actor at 0x7f984c210c20>",
-        "make_critic": "<function SACPolicy.make_critic at 0x7f984c210cc0>",
-        "forward": "<function SACPolicy.forward at 0x7f984c210d60>",
-        "_predict": "<function SACPolicy._predict at 0x7f984c210e00>",
-        "set_training_mode": "<function SACPolicy.set_training_mode at 0x7f984c210ea0>",
         "__abstractmethods__": "frozenset()",
-        "_abc_impl": "<_abc._abc_data object at 0x7f984c21ca00>"
     },
     "verbose": 0,
     "policy_kwargs": {
         "use_sde": false
     },
-    "num_timesteps": 2000000,
-    "_total_timesteps": 2000000,
     "_num_timesteps_at_start": 0,
     "seed": 0,
     "action_noise": null,
-    "start_time": 1729236929340004656,
     "learning_rate": 0.0003,
-    "tensorboard_log": "runs/q8ptehk8",
     "_last_obs": {
         ":type:": "<class 'numpy.ndarray'>",
-        ":serialized:": "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"
     },
     "_last_episode_starts": {
         ":type:": "<class 'numpy.ndarray'>",
@@ -39,22 +39,22 @@
     },
     "_last_original_obs": {
         ":type:": "<class 'numpy.ndarray'>",
-        ":serialized:": "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"
     },
-    "_episode_num": 6392,
     "use_sde": false,
     "sde_sample_freq": -1,
     "_current_progress_remaining": 0.0,
     "_stats_window_size": 100,
     "ep_info_buffer": {
         ":type:": "<class 'collections.deque'>",
-        ":serialized:": "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"
     },
     "ep_success_buffer": {
         ":type:": "<class 'collections.deque'>",
         ":serialized:": "gAWVIAAAAAAAAACMC2NvbGxlY3Rpb25zlIwFZGVxdWWUk5QpS2SGlFKULg=="
     },
-    "_n_updates": 398000,
     "buffer_size": 1000000,
     "batch_size": 256,
     "learning_starts": 10000,
@@ -68,13 +68,13 @@
         "__module__": "stable_baselines3.common.buffers",
         "__annotations__": "{'observations': <class 'numpy.ndarray'>, 'next_observations': <class 'numpy.ndarray'>, 'actions': <class 'numpy.ndarray'>, 'rewards': <class 'numpy.ndarray'>, 'dones': <class 'numpy.ndarray'>, 'timeouts': <class 'numpy.ndarray'>}",
         "__doc__": "\n    Replay buffer used in off-policy algorithms like SAC/TD3.\n\n    :param buffer_size: Max number of element in the buffer\n    :param observation_space: Observation space\n    :param action_space: Action space\n    :param device: PyTorch device\n    :param n_envs: Number of parallel environments\n    :param optimize_memory_usage: Enable a memory efficient variant\n        of the replay buffer which reduces by almost a factor two the memory used,\n        at a cost of more complexity.\n        See https://github.com/DLR-RM/stable-baselines3/issues/37#issuecomment-637501195\n        and https://github.com/DLR-RM/stable-baselines3/pull/28#issuecomment-637559274\n        Cannot be used in combination with handle_timeout_termination.\n    :param handle_timeout_termination: Handle timeout termination (due to timelimit)\n        separately and treat the task as infinite horizon task.\n        https://github.com/DLR-RM/stable-baselines3/issues/284\n    ",
-        "__init__": "<function ReplayBuffer.__init__ at 0x7f9857060c20>",
-        "add": "<function ReplayBuffer.add at 0x7f9857060d60>",
-        "sample": "<function ReplayBuffer.sample at 0x7f9857060e00>",
-        "_get_samples": "<function ReplayBuffer._get_samples at 0x7f9857060ea0>",
-        "_maybe_cast_dtype": "<staticmethod(<function ReplayBuffer._maybe_cast_dtype at 0x7f9857060f40>)>",
         "__abstractmethods__": "frozenset()",
-        "_abc_impl": "<_abc._abc_data object at 0x7f985705a680>"
     },
     "replay_buffer_kwargs": {},
     "train_freq": {

         "__module__": "stable_baselines3.sac.policies",
         "__annotations__": "{'actor': <class 'stable_baselines3.sac.policies.Actor'>, 'critic': <class 'stable_baselines3.common.policies.ContinuousCritic'>, 'critic_target': <class 'stable_baselines3.common.policies.ContinuousCritic'>}",
         "__doc__": "\n    Policy class (with both actor and critic) for SAC.\n\n    :param observation_space: Observation space\n    :param action_space: Action space\n    :param lr_schedule: Learning rate schedule (could be constant)\n    :param net_arch: The specification of the policy and value networks.\n    :param activation_fn: Activation function\n    :param use_sde: Whether to use State Dependent Exploration or not\n    :param log_std_init: Initial value for the log standard deviation\n    :param use_expln: Use ``expln()`` function instead of ``exp()`` when using gSDE to ensure\n        a positive standard deviation (cf paper). It allows to keep variance\n        above zero and prevent it from growing too fast. In practice, ``exp()`` is usually enough.\n    :param clip_mean: Clip the mean output when using gSDE to avoid numerical instability.\n    :param features_extractor_class: Features extractor to use.\n    :param features_extractor_kwargs: Keyword arguments\n        to pass to the features extractor.\n    :param normalize_images: Whether to normalize images or not,\n         dividing by 255.0 (True by default)\n    :param optimizer_class: The optimizer to use,\n        ``th.optim.Adam`` by default\n    :param optimizer_kwargs: Additional keyword arguments,\n        excluding the learning rate, to pass to the optimizer\n    :param n_critics: Number of critic networks to create.\n    :param share_features_extractor: Whether to share or not the features extractor\n        between the actor and the critic (this saves computation time)\n    ",
+        "__init__": "<function SACPolicy.__init__ at 0x7ff9dcc04400>",
+        "_build": "<function SACPolicy._build at 0x7ff9dcc04a40>",
+        "_get_constructor_parameters": "<function SACPolicy._get_constructor_parameters at 0x7ff9dcc04ae0>",
+        "reset_noise": "<function SACPolicy.reset_noise at 0x7ff9dcc04b80>",
+        "make_actor": "<function SACPolicy.make_actor at 0x7ff9dcc04c20>",
+        "make_critic": "<function SACPolicy.make_critic at 0x7ff9dcc04cc0>",
+        "forward": "<function SACPolicy.forward at 0x7ff9dcc04d60>",
+        "_predict": "<function SACPolicy._predict at 0x7ff9dcc04e00>",
+        "set_training_mode": "<function SACPolicy.set_training_mode at 0x7ff9dcc04ea0>",
         "__abstractmethods__": "frozenset()",
+        "_abc_impl": "<_abc._abc_data object at 0x7ff9dcc0cd80>"
     },
     "verbose": 0,
     "policy_kwargs": {
         "use_sde": false
     },
+    "num_timesteps": 10000000,
+    "_total_timesteps": 10000000,
     "_num_timesteps_at_start": 0,
     "seed": 0,
     "action_noise": null,
+    "start_time": 1729923137747710094,
     "learning_rate": 0.0003,
+    "tensorboard_log": "runs/brmzsgdj",
     "_last_obs": {
         ":type:": "<class 'numpy.ndarray'>",
+        ":serialized:": "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"
     },
     "_last_episode_starts": {
         ":type:": "<class 'numpy.ndarray'>",
     },
     "_last_original_obs": {
         ":type:": "<class 'numpy.ndarray'>",
+        ":serialized:": "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"
     },
+    "_episode_num": 14991,
     "use_sde": false,
     "sde_sample_freq": -1,
     "_current_progress_remaining": 0.0,
     "_stats_window_size": 100,
     "ep_info_buffer": {
         ":type:": "<class 'collections.deque'>",
+        ":serialized:": "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"
     },
     "ep_success_buffer": {
         ":type:": "<class 'collections.deque'>",
         ":serialized:": "gAWVIAAAAAAAAACMC2NvbGxlY3Rpb25zlIwFZGVxdWWUk5QpS2SGlFKULg=="
     },
+    "_n_updates": 1998000,
     "buffer_size": 1000000,
     "batch_size": 256,
     "learning_starts": 10000,
         "__module__": "stable_baselines3.common.buffers",
         "__annotations__": "{'observations': <class 'numpy.ndarray'>, 'next_observations': <class 'numpy.ndarray'>, 'actions': <class 'numpy.ndarray'>, 'rewards': <class 'numpy.ndarray'>, 'dones': <class 'numpy.ndarray'>, 'timeouts': <class 'numpy.ndarray'>}",
         "__doc__": "\n    Replay buffer used in off-policy algorithms like SAC/TD3.\n\n    :param buffer_size: Max number of element in the buffer\n    :param observation_space: Observation space\n    :param action_space: Action space\n    :param device: PyTorch device\n    :param n_envs: Number of parallel environments\n    :param optimize_memory_usage: Enable a memory efficient variant\n        of the replay buffer which reduces by almost a factor two the memory used,\n        at a cost of more complexity.\n        See https://github.com/DLR-RM/stable-baselines3/issues/37#issuecomment-637501195\n        and https://github.com/DLR-RM/stable-baselines3/pull/28#issuecomment-637559274\n        Cannot be used in combination with handle_timeout_termination.\n    :param handle_timeout_termination: Handle timeout termination (due to timelimit)\n        separately and treat the task as infinite horizon task.\n        https://github.com/DLR-RM/stable-baselines3/issues/284\n    ",
+        "__init__": "<function ReplayBuffer.__init__ at 0x7ff9e7ae8c20>",
+        "add": "<function ReplayBuffer.add at 0x7ff9e7ae8d60>",
+        "sample": "<function ReplayBuffer.sample at 0x7ff9e7ae8e00>",
+        "_get_samples": "<function ReplayBuffer._get_samples at 0x7ff9e7ae8ea0>",
+        "_maybe_cast_dtype": "<staticmethod(<function ReplayBuffer._maybe_cast_dtype at 0x7ff9e7ae8f40>)>",
         "__abstractmethods__": "frozenset()",
+        "_abc_impl": "<_abc._abc_data object at 0x7ff9e7ade1c0>"
     },
     "replay_buffer_kwargs": {},
     "train_freq": {

humanoid-v5-sac-medium/ent_coef_optimizer.pth CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:117d4e619d79c7c1defaf4be8c4c0a8b8f41fd579830f051ae79a12732ebf684
 size 1940

 version https://git-lfs.github.com/spec/v1
+oid sha256:0a5eb9a761957cf046a62e7f0443022dcdcb9be5b5f2cfb9cf2e9246d957f16f
 size 1940

humanoid-v5-sac-medium/policy.pth CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35cd49a0f802b6b301bba08b832298930247139315e782d9c53d1892d396da9b
 size 3222902

 version https://git-lfs.github.com/spec/v1
+oid sha256:a5309b8aa292296cb70c71b1c1c9223a4974ba8f2a89f975937cff997594712f
 size 3222902

humanoid-v5-sac-medium/pytorch_variables.pth CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a6541940212b3b0637e21107f51e8dee8727e586cf766b073957e032a092b2af
 size 1180

 version https://git-lfs.github.com/spec/v1
+oid sha256:c261837f5d1b77f7d2eba152303ae8c3bae57048ebb25ece1cb564872652c039
 size 1180

replay.mp4 CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:56fdc794bae27f0f65de65cf0cd3ad5e965b67adaf2c43f6a6341cb63ba6db62
-size 1469616

 version https://git-lfs.github.com/spec/v1
+oid sha256:ab332911fb5a785fe86675c30eb6bae2f75eb2c5d5137e4288fe8300f1658f98
+size 1382864

results.json CHANGED Viewed

	@@ -1 +1 @@
1	- {"mean_reward": ~~4731~~.~~3818424~~, "std_reward": ~~1469~~.~~856932809794~~, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2024-10-~~18T13~~:07:21.~~666264~~"}


1	+ {"mean_reward": 6846.050477999999, "std_reward": 18.344659603799315, "is_deterministic": true, "n_eval_episodes": 10, "eval_datetime": "2024-10-26T17:08:38.081409"}