Created
September 4, 2020 02:24
-
-
Save krzentner/1b30f48286377908751da511f266544f to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| { | |
| "$type": "garage.experiment.local_runner.LocalRunner", | |
| "_algo": { | |
| "$type": "garage.torch.algos.pearl.PEARL", | |
| "_batch_size": 256, | |
| "_context_replay_buffers": { | |
| "0": { | |
| "$type": "garage.replay_buffer.path_buffer.PathBuffer", | |
| "_buffer": {}, | |
| "_capacity": 1000000, | |
| "_first_idx_of_next_path": 0, | |
| "_path_segments": [], | |
| "_transitions_stored": 0 | |
| }, | |
| "1": { | |
| "$type": "garage.replay_buffer.path_buffer.PathBuffer", | |
| "_buffer": {}, | |
| "_capacity": 1000000, | |
| "_first_idx_of_next_path": 0, | |
| "_path_segments": [], | |
| "_transitions_stored": 0 | |
| } | |
| }, | |
| "_discount": 0.99, | |
| "_embedding_batch_size": 100, | |
| "_embedding_mini_batch_size": 100, | |
| "_env": [ | |
| { | |
| "$type": "garage.sampler.env_update.SetTaskUpdate", | |
| "_env_constructor": { | |
| "$type": "builtins.function" | |
| }, | |
| "_task": { | |
| "direction": -1.0 | |
| } | |
| }, | |
| { | |
| "$type": "garage.sampler.env_update.SetTaskUpdate", | |
| "_env_constructor": { | |
| "$type": "builtins.function" | |
| }, | |
| "_task": { | |
| "direction": -1.0 | |
| } | |
| } | |
| ], | |
| "_evaluator": { | |
| "$type": "garage.experiment.meta_evaluator.MetaEvaluator", | |
| "_eval_itr": 0, | |
| "_max_episode_length": 200, | |
| "_n_exploration_eps": 10, | |
| "_n_test_episodes": 1, | |
| "_n_test_tasks": 2, | |
| "_prefix": "MetaTest", | |
| "_test_sampler": null, | |
| "_test_task_names": null, | |
| "_test_task_sampler": { | |
| "$type": "garage.experiment.task_sampler.SetTaskSampler", | |
| "_env": { | |
| "$type": "garage.envs.normalized_env.NormalizedEnv", | |
| "_env": { | |
| "$type": "garage.envs.gym_env.GymEnv", | |
| "_action_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669BBA0", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_env": { | |
| "$type": "garage.envs.mujoco.half_cheetah_dir_env.HalfCheetahDirEnv", | |
| "_ezpickle_args": [], | |
| "_ezpickle_kwargs": {}, | |
| "_task": { | |
| "direction": 1.0 | |
| }, | |
| "_viewers": {}, | |
| "action_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669B888", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "data": "<mujoco_py.cymj.PyMjData object at 0x79f0118>", | |
| "frame_skip": 5, | |
| "init_qpos": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "init_qvel": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "metadata": { | |
| "render.modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "video.frames_per_second": 20 | |
| }, | |
| "model": "<mujoco_py.cymj.PyMjModel object at 0x7be3bf8>", | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669BA98", | |
| "observation_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669B990", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "sim": "<mujoco_py.cymj.MjSim object at 0x7eff064da378>", | |
| "spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669B888", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669B990", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "viewer": null | |
| }, | |
| "_max_episode_length": null, | |
| "_observation_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669BCA8", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "_render_modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669BBA0", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669BCA8", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "_step_cnt": null, | |
| "_visualize": false | |
| }, | |
| "_expected_action_scale": 1.0, | |
| "_flatten_obs": true, | |
| "_normalize_obs": false, | |
| "_normalize_reward": false, | |
| "_obs_alpha": 0.001, | |
| "_obs_mean": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "_obs_var": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "_reward_alpha": 0.001, | |
| "_reward_mean": 0.0, | |
| "_reward_var": 1.0, | |
| "_scale_reward": 1.0 | |
| }, | |
| "_env_constructor": { | |
| "$type": "builtins.function" | |
| } | |
| }, | |
| "_worker_args": { | |
| "accum_context": true, | |
| "deterministic": true | |
| }, | |
| "_worker_class": { | |
| "$typename": "garage.torch.algos.pearl.PEARLWorker" | |
| } | |
| }, | |
| "_is_resuming": false, | |
| "_kl_lambda": 0.1, | |
| "_latent_dim": 5, | |
| "_meta_batch_size": 2, | |
| "_num_extra_rl_steps_posterior": 30000, | |
| "_num_initial_steps": 100000, | |
| "_num_steps_per_epoch": 100000, | |
| "_num_steps_posterior": 0, | |
| "_num_steps_prior": 20000, | |
| "_num_tasks_sample": 2, | |
| "_num_test_tasks": 2, | |
| "_num_train_tasks": 2, | |
| "_policy": { | |
| "$type": "garage.torch.policies.context_conditioned_policy.ContextConditionedPolicy", | |
| "_backward_hooks": {}, | |
| "_buffers": { | |
| "z": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "z_means": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "z_vars": { | |
| "$type": "torch.Tensor" | |
| } | |
| }, | |
| "_context": null, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_latent_dim": 5, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_context_encoder": { | |
| "$type": "garage.torch.embeddings.mlp_encoder.MLPEncoder", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 27, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 10, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_output_dim": 10, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_policy": { | |
| "$type": "garage.torch.policies.tanh_gaussian_mlp_policy.TanhGaussianMLPPolicy", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_env_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338468", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338360", | |
| "shape": [ | |
| 25 | |
| ] | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_module": { | |
| "$type": "garage.torch.modules.gaussian_mlp_module.GaussianMLPTwoHeadedModule", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": { | |
| "max_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "min_std_param": { | |
| "$type": "torch.Tensor" | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_hidden_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_hidden_nonlinearity": { | |
| "$typename": "torch.nn.modules.activation.ReLU" | |
| }, | |
| "_hidden_sizes": [ | |
| 300, | |
| 300, | |
| 300 | |
| ], | |
| "_hidden_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_input_dim": 25.0, | |
| "_layer_normalization": false, | |
| "_learn_std": true, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_max_std": 7.38905609893065, | |
| "_max_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "_min_std": 2.061153622438558e-09, | |
| "_min_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "_modules": { | |
| "_shared_mean_log_std_network": { | |
| "$type": "garage.torch.modules.multi_headed_mlp_module.MultiHeadedMLPModule", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 25.0, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 6.0, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 6.0, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_norm_dist_class": { | |
| "$typename": "garage.torch.distributions.tanh_normal.TanhNormal" | |
| }, | |
| "_output_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_output_nonlinearity": null, | |
| "_output_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_parameters": { | |
| "_init_std": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "_std_hidden_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_hidden_nonlinearity": "<built-in method tanh of type object at 0x7eff9e9de860>", | |
| "_std_hidden_sizes": [ | |
| 32, | |
| 32 | |
| ], | |
| "_std_hidden_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_output_nonlinearity": null, | |
| "_std_output_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_parameterization": "exp", | |
| "training": true | |
| } | |
| }, | |
| "_name": "TanhGaussianPolicy", | |
| "_obs_dim": 25.0, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "_use_information_bottleneck": true, | |
| "_use_next_obs": false, | |
| "training": true | |
| }, | |
| "_policy_mean_reg_coeff": 0.001, | |
| "_policy_optimizer": { | |
| "$type": "torch.optim.adam.Adam", | |
| "defaults": { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "weight_decay": 0 | |
| }, | |
| "param_groups": [ | |
| { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "params": [ | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| ], | |
| "weight_decay": 0 | |
| } | |
| ], | |
| "state": {} | |
| }, | |
| "_policy_pre_activation_coeff": 0.0, | |
| "_policy_std_reg_coeff": 0.001, | |
| "_qf1": { | |
| "$type": "garage.torch.q_functions.continuous_mlp_q_function.ContinuousMLPQFunction", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_env_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338468", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338360", | |
| "shape": [ | |
| 25 | |
| ] | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 31.0, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 1, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_obs_dim": 25.0, | |
| "_output_dim": 1, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_qf2": { | |
| "$type": "garage.torch.q_functions.continuous_mlp_q_function.ContinuousMLPQFunction", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_env_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669BDB0", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF0669BEB8", | |
| "shape": [ | |
| 25 | |
| ] | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 31.0, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 1, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_obs_dim": 25.0, | |
| "_output_dim": 1, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_replay_buffer_size": 1000000, | |
| "_replay_buffers": { | |
| "0": { | |
| "$type": "garage.replay_buffer.path_buffer.PathBuffer", | |
| "_buffer": {}, | |
| "_capacity": 1000000, | |
| "_first_idx_of_next_path": 0, | |
| "_path_segments": [], | |
| "_transitions_stored": 0 | |
| }, | |
| "1": { | |
| "$type": "garage.replay_buffer.path_buffer.PathBuffer", | |
| "_buffer": {}, | |
| "_capacity": 1000000, | |
| "_first_idx_of_next_path": 0, | |
| "_path_segments": [], | |
| "_transitions_stored": 0 | |
| } | |
| }, | |
| "_reward_scale": 5.0, | |
| "_soft_target_tau": 0.005, | |
| "_task_idx": null, | |
| "_update_post_train": 1, | |
| "_use_information_bottleneck": true, | |
| "_use_next_obs_in_context": false, | |
| "_vf": { | |
| "$type": "garage.torch.q_functions.continuous_mlp_q_function.ContinuousMLPQFunction", | |
| "_action_dim": 5.0, | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_env_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338678", | |
| "shape": [ | |
| 5 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338570", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 25.0, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 1, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_obs_dim": 20.0, | |
| "_output_dim": 1, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "context_optimizer": { | |
| "$type": "torch.optim.adam.Adam", | |
| "defaults": { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "weight_decay": 0 | |
| }, | |
| "param_groups": [ | |
| { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "params": [ | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| ], | |
| "weight_decay": 0 | |
| } | |
| ], | |
| "state": {} | |
| }, | |
| "max_episode_length": 200, | |
| "qf1_optimizer": { | |
| "$type": "torch.optim.adam.Adam", | |
| "defaults": { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "weight_decay": 0 | |
| }, | |
| "param_groups": [ | |
| { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "params": [ | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| ], | |
| "weight_decay": 0 | |
| } | |
| ], | |
| "state": {} | |
| }, | |
| "qf2_optimizer": { | |
| "$type": "torch.optim.adam.Adam", | |
| "defaults": { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "weight_decay": 0 | |
| }, | |
| "param_groups": [ | |
| { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "params": [ | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| ], | |
| "weight_decay": 0 | |
| } | |
| ], | |
| "state": {} | |
| }, | |
| "target_vf": { | |
| "$type": "garage.torch.q_functions.continuous_mlp_q_function.ContinuousMLPQFunction", | |
| "_action_dim": 5.0, | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_env_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338888", | |
| "shape": [ | |
| 5 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338780", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 25.0, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 1, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_obs_dim": 20.0, | |
| "_output_dim": 1, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "vf_criterion": { | |
| "$type": "torch.nn.modules.loss.MSELoss", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "reduction": "mean", | |
| "training": true | |
| }, | |
| "vf_optimizer": { | |
| "$type": "torch.optim.adam.Adam", | |
| "defaults": { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "weight_decay": 0 | |
| }, | |
| "param_groups": [ | |
| { | |
| "amsgrad": false, | |
| "betas": [ | |
| 0.9, | |
| 0.999 | |
| ], | |
| "eps": 1e-08, | |
| "lr": 0.0003, | |
| "params": [ | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| ], | |
| "weight_decay": 0 | |
| } | |
| ], | |
| "state": {} | |
| } | |
| }, | |
| "_env": { | |
| "$type": "garage.envs.normalized_env.NormalizedEnv", | |
| "_env": { | |
| "$type": "garage.envs.gym_env.GymEnv", | |
| "_action_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338048", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_env": { | |
| "$type": "garage.envs.mujoco.half_cheetah_dir_env.HalfCheetahDirEnv", | |
| "_ezpickle_args": [], | |
| "_ezpickle_kwargs": {}, | |
| "_task": { | |
| "direction": -1.0 | |
| }, | |
| "_viewers": {}, | |
| "action_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338A98", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "data": "<mujoco_py.cymj.PyMjData object at 0x7bfd778>", | |
| "frame_skip": 5, | |
| "init_qpos": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "init_qvel": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "metadata": { | |
| "render.modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "video.frames_per_second": 20 | |
| }, | |
| "model": "<mujoco_py.cymj.PyMjModel object at 0x94cc838>", | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338150", | |
| "observation_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338990", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "sim": "<mujoco_py.cymj.MjSim object at 0x7eff062d3ea0>", | |
| "spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338A98", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338990", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "viewer": null | |
| }, | |
| "_max_episode_length": null, | |
| "_observation_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338258", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "_render_modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338048", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338258", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "_step_cnt": null, | |
| "_visualize": false | |
| }, | |
| "_expected_action_scale": 1.0, | |
| "_flatten_obs": true, | |
| "_normalize_obs": false, | |
| "_normalize_reward": false, | |
| "_obs_alpha": 0.001, | |
| "_obs_mean": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "_obs_var": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "_reward_alpha": 0.001, | |
| "_reward_mean": 0.0, | |
| "_reward_var": 1.0, | |
| "_scale_reward": 1.0 | |
| }, | |
| "_has_setup": true, | |
| "_itr_start_time": null, | |
| "_n_workers": 1, | |
| "_plot": false, | |
| "_plotter": null, | |
| "_sampler": { | |
| "$type": "garage.sampler.local_sampler.LocalSampler", | |
| "_agents": [ | |
| { | |
| "$type": "garage.torch.policies.context_conditioned_policy.ContextConditionedPolicy", | |
| "_backward_hooks": {}, | |
| "_buffers": { | |
| "z": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "z_means": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "z_vars": { | |
| "$type": "torch.Tensor" | |
| } | |
| }, | |
| "_context": null, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_latent_dim": 5, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_context_encoder": { | |
| "$type": "garage.torch.embeddings.mlp_encoder.MLPEncoder", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 27, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 10, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_output_dim": 10, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_policy": { | |
| "$type": "garage.torch.policies.tanh_gaussian_mlp_policy.TanhGaussianMLPPolicy", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_env_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338468", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338360", | |
| "shape": [ | |
| 25 | |
| ] | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_module": { | |
| "$type": "garage.torch.modules.gaussian_mlp_module.GaussianMLPTwoHeadedModule", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": { | |
| "max_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "min_std_param": { | |
| "$type": "torch.Tensor" | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_hidden_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_hidden_nonlinearity": { | |
| "$typename": "torch.nn.modules.activation.ReLU" | |
| }, | |
| "_hidden_sizes": [ | |
| 300, | |
| 300, | |
| 300 | |
| ], | |
| "_hidden_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_input_dim": 25.0, | |
| "_layer_normalization": false, | |
| "_learn_std": true, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_max_std": 7.38905609893065, | |
| "_max_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "_min_std": 2.061153622438558e-09, | |
| "_min_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "_modules": { | |
| "_shared_mean_log_std_network": { | |
| "$type": "garage.torch.modules.multi_headed_mlp_module.MultiHeadedMLPModule", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 25.0, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 6.0, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 6.0, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_norm_dist_class": { | |
| "$typename": "garage.torch.distributions.tanh_normal.TanhNormal" | |
| }, | |
| "_output_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_output_nonlinearity": null, | |
| "_output_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_parameters": { | |
| "_init_std": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "_std_hidden_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_hidden_nonlinearity": "<built-in method tanh of type object at 0x7eff9e9de860>", | |
| "_std_hidden_sizes": [ | |
| 32, | |
| 32 | |
| ], | |
| "_std_hidden_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_output_nonlinearity": null, | |
| "_std_output_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_parameterization": "exp", | |
| "training": true | |
| } | |
| }, | |
| "_name": "TanhGaussianPolicy", | |
| "_obs_dim": 25.0, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "_use_information_bottleneck": true, | |
| "_use_next_obs": false, | |
| "training": true | |
| } | |
| ], | |
| "_envs": [ | |
| { | |
| "$type": "garage.envs.normalized_env.NormalizedEnv", | |
| "_env": { | |
| "$type": "garage.envs.gym_env.GymEnv", | |
| "_action_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56678", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_env": { | |
| "$type": "garage.envs.mujoco.half_cheetah_dir_env.HalfCheetahDirEnv", | |
| "_ezpickle_args": [], | |
| "_ezpickle_kwargs": {}, | |
| "_task": { | |
| "direction": -1.0 | |
| }, | |
| "_viewers": {}, | |
| "action_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56150", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "data": "<mujoco_py.cymj.PyMjData object at 0x6eb2a88>", | |
| "frame_skip": 5, | |
| "init_qpos": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "init_qvel": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "metadata": { | |
| "render.modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "video.frames_per_second": 20 | |
| }, | |
| "model": "<mujoco_py.cymj.PyMjModel object at 0x82187108>", | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56360", | |
| "observation_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56258", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "sim": "<mujoco_py.cymj.MjSim object at 0x7efef9d0b2f0>", | |
| "spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56150", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56258", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "viewer": null | |
| }, | |
| "_max_episode_length": null, | |
| "_observation_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56780", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "_render_modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56678", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56780", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "_step_cnt": null, | |
| "_visualize": false | |
| }, | |
| "_expected_action_scale": 1.0, | |
| "_flatten_obs": true, | |
| "_normalize_obs": false, | |
| "_normalize_reward": false, | |
| "_obs_alpha": 0.001, | |
| "_obs_mean": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "_obs_var": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "_reward_alpha": 0.001, | |
| "_reward_mean": 0.0, | |
| "_reward_var": 1.0, | |
| "_scale_reward": 1.0 | |
| } | |
| ], | |
| "_factory": { | |
| "$type": "garage.sampler.worker_factory.WorkerFactory", | |
| "_max_episode_length": 200, | |
| "_seed": 10000, | |
| "_worker_args": {}, | |
| "_worker_class": { | |
| "$typename": "garage.torch.algos.pearl.PEARLWorker" | |
| }, | |
| "n_workers": 1 | |
| }, | |
| "_workers": [ | |
| { | |
| "$type": "garage.torch.algos.pearl.PEARLWorker", | |
| "_accum_context": false, | |
| "_agent_infos": {}, | |
| "_deterministic": false, | |
| "_env_steps": [], | |
| "_eps_length": 0, | |
| "_last_observations": [], | |
| "_lengths": [], | |
| "_max_episode_length": 200, | |
| "_observations": [], | |
| "_prev_obs": null, | |
| "_seed": 10000, | |
| "_worker_number": 0, | |
| "agent": { | |
| "$type": "garage.torch.policies.context_conditioned_policy.ContextConditionedPolicy", | |
| "_backward_hooks": {}, | |
| "_buffers": { | |
| "z": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "z_means": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "z_vars": { | |
| "$type": "torch.Tensor" | |
| } | |
| }, | |
| "_context": null, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_latent_dim": 5, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_context_encoder": { | |
| "$type": "garage.torch.embeddings.mlp_encoder.MLPEncoder", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 27, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 200, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "module": { | |
| "$type": "builtins.function" | |
| }, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 200, | |
| "out_features": 10, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_output_dim": 10, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_policy": { | |
| "$type": "garage.torch.policies.tanh_gaussian_mlp_policy.TanhGaussianMLPPolicy", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_env_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338468", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFF06338360", | |
| "shape": [ | |
| 25 | |
| ] | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_module": { | |
| "$type": "garage.torch.modules.gaussian_mlp_module.GaussianMLPTwoHeadedModule", | |
| "_action_dim": 6.0, | |
| "_backward_hooks": {}, | |
| "_buffers": { | |
| "max_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "min_std_param": { | |
| "$type": "torch.Tensor" | |
| } | |
| }, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_hidden_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_hidden_nonlinearity": { | |
| "$typename": "torch.nn.modules.activation.ReLU" | |
| }, | |
| "_hidden_sizes": [ | |
| 300, | |
| 300, | |
| 300 | |
| ], | |
| "_hidden_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_input_dim": 25.0, | |
| "_layer_normalization": false, | |
| "_learn_std": true, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_max_std": 7.38905609893065, | |
| "_max_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "_min_std": 2.061153622438558e-09, | |
| "_min_std_param": { | |
| "$type": "torch.Tensor" | |
| }, | |
| "_modules": { | |
| "_shared_mean_log_std_network": { | |
| "$type": "garage.torch.modules.multi_headed_mlp_module.MultiHeadedMLPModule", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 25.0, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "2": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 300, | |
| "training": true | |
| }, | |
| "non_linearity": { | |
| "$type": "garage.torch._functions.NonLinearity", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "module": { | |
| "$type": "torch.nn.modules.activation.ReLU", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "inplace": false, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "_output_layers": { | |
| "$type": "torch.nn.modules.container.ModuleList", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "0": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 6.0, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| }, | |
| "1": { | |
| "$type": "torch.nn.modules.container.Sequential", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": { | |
| "linear": { | |
| "$type": "torch.nn.modules.linear.Linear", | |
| "_backward_hooks": {}, | |
| "_buffers": {}, | |
| "_forward_hooks": {}, | |
| "_forward_pre_hooks": {}, | |
| "_load_state_dict_pre_hooks": {}, | |
| "_modules": {}, | |
| "_parameters": { | |
| "bias": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| }, | |
| "weight": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "in_features": 300, | |
| "out_features": 6.0, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_norm_dist_class": { | |
| "$typename": "garage.torch.distributions.tanh_normal.TanhNormal" | |
| }, | |
| "_output_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_output_nonlinearity": null, | |
| "_output_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_parameters": { | |
| "_init_std": { | |
| "$type": "torch.nn.parameter.Parameter" | |
| } | |
| }, | |
| "_state_dict_hooks": {}, | |
| "_std_hidden_b_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_hidden_nonlinearity": "<built-in method tanh of type object at 0x7eff9e9de860>", | |
| "_std_hidden_sizes": [ | |
| 32, | |
| 32 | |
| ], | |
| "_std_hidden_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_output_nonlinearity": null, | |
| "_std_output_w_init": { | |
| "$type": "builtins.function" | |
| }, | |
| "_std_parameterization": "exp", | |
| "training": true | |
| } | |
| }, | |
| "_name": "TanhGaussianPolicy", | |
| "_obs_dim": 25.0, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "training": true | |
| } | |
| }, | |
| "_parameters": {}, | |
| "_state_dict_hooks": {}, | |
| "_use_information_bottleneck": true, | |
| "_use_next_obs": false, | |
| "training": true | |
| }, | |
| "env": { | |
| "$type": "garage.envs.normalized_env.NormalizedEnv", | |
| "_env": { | |
| "$type": "garage.envs.gym_env.GymEnv", | |
| "_action_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56678", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_env": { | |
| "$type": "garage.envs.mujoco.half_cheetah_dir_env.HalfCheetahDirEnv", | |
| "_ezpickle_args": [], | |
| "_ezpickle_kwargs": {}, | |
| "_task": { | |
| "direction": -1.0 | |
| }, | |
| "_viewers": {}, | |
| "action_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56150", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "data": "<mujoco_py.cymj.PyMjData object at 0x6eb2a88>", | |
| "frame_skip": 5, | |
| "init_qpos": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "init_qvel": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "metadata": { | |
| "render.modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "video.frames_per_second": 20 | |
| }, | |
| "model": "<mujoco_py.cymj.PyMjModel object at 0x82187108>", | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56360", | |
| "observation_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56258", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "sim": "<mujoco_py.cymj.MjSim object at 0x7efef9d0b2f0>", | |
| "spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56150", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "gym.spaces.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56258", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "viewer": null | |
| }, | |
| "_max_episode_length": null, | |
| "_observation_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56780", | |
| "shape": [ | |
| 20 | |
| ] | |
| }, | |
| "_render_modes": [ | |
| "human", | |
| "rgb_array", | |
| "depth_array" | |
| ], | |
| "_spec": { | |
| "$type": "garage._environment.EnvSpec", | |
| "_input_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "bounded_below": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "low": [ | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0, | |
| -1.0 | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56678", | |
| "shape": [ | |
| 6 | |
| ] | |
| }, | |
| "_max_episode_length": null, | |
| "_output_space": { | |
| "$type": "akro.box.Box", | |
| "bounded_above": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "bounded_below": [ | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false, | |
| false | |
| ], | |
| "dtype": "dtype('float32')", | |
| "high": [ | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity, | |
| Infinity | |
| ], | |
| "low": [ | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity, | |
| -Infinity | |
| ], | |
| "np_random": "RandomState(MT19937) at 0x7EFEF9D56780", | |
| "shape": [ | |
| 20 | |
| ] | |
| } | |
| }, | |
| "_step_cnt": null, | |
| "_visualize": false | |
| }, | |
| "_expected_action_scale": 1.0, | |
| "_flatten_obs": true, | |
| "_normalize_obs": false, | |
| "_normalize_reward": false, | |
| "_obs_alpha": 0.001, | |
| "_obs_mean": [ | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0, | |
| 0.0 | |
| ], | |
| "_obs_var": [ | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0, | |
| 1.0 | |
| ], | |
| "_reward_alpha": 0.001, | |
| "_reward_mean": 0.0, | |
| "_reward_var": 1.0, | |
| "_scale_reward": 1.0 | |
| } | |
| } | |
| ] | |
| }, | |
| "_setup_args": { | |
| "$type": "garage.experiment.local_runner.SetupArgs", | |
| "sampler_args": { | |
| "max_episode_length": 200 | |
| }, | |
| "sampler_cls": { | |
| "$typename": "garage.sampler.local_sampler.LocalSampler" | |
| }, | |
| "seed": 10000 | |
| }, | |
| "_snapshotter": { | |
| "$type": "garage.experiment.snapshotter.Snapshotter", | |
| "_snapshot_dir": "/home/kr/garage/data/local/experiment/old_pearl_half_cheetah_dir_25", | |
| "_snapshot_gap": 1, | |
| "_snapshot_mode": "last" | |
| }, | |
| "_start_time": null, | |
| "_stats": { | |
| "$type": "garage.experiment.local_runner.ExperimentStats", | |
| "last_episode": null, | |
| "total_env_steps": 0, | |
| "total_epoch": 0, | |
| "total_itr": 0 | |
| }, | |
| "_train_args": { | |
| "$type": "garage.experiment.local_runner.TrainArgs", | |
| "batch_size": 256, | |
| "n_epochs": 500, | |
| "pause_for_plot": false, | |
| "plot": false, | |
| "start_epoch": 0, | |
| "store_episodes": false | |
| }, | |
| "_worker_args": {}, | |
| "_worker_class": { | |
| "$typename": "garage.torch.algos.pearl.PEARLWorker" | |
| }, | |
| "enable_logging": true, | |
| "step_episode": null, | |
| "step_itr": null | |
| } |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment