Explore Results

{
  "stop": {
    "timesteps_total": 5000000,
    "episode_len_mean": 300,
    "training_iteration": 1000000,
    "episode_reward_mean": 50
  },
  "config": {
    "lr": 2e-05,
    "env": "knights_archers_zombies_v10",
    "seed": null,
    "gamma": 0.99,
    "input": "sampler",
    "model": {
      "dim": 84,
      "use_lstm": false,
      "grayscale": false,
      "zero_mean": true,
      "framestack": true,
      "_time_major": false,
      "max_seq_len": 20,
      "conv_filters": null,
      "custom_model": null,
      "free_log_std": false,
      "attention_dim": 64,
      "fcnet_hiddens": [
        256,
        256
      ],
      "use_attention": false,
      "lstm_cell_size": 256,
      "conv_activation": "relu",
      "no_final_linear": false,
      "vf_share_layers": false,
      "fcnet_activation": "tanh",
      "attention_head_dim": 32,
      "custom_action_dist": null,
      "encoder_latent_dim": null,
      "post_fcnet_hiddens": [],
      "always_check_shapes": false,
      "attention_num_heads": 1,
      "custom_model_config": {},
      "custom_preprocessor": null,
      "lstm_use_prev_action": false,
      "lstm_use_prev_reward": false,
      "conv_bias_initializer": null,
      "lstm_bias_initializer": null,
      "post_fcnet_activation": "relu",
      "fcnet_bias_initializer": null,
      "conv_kernel_initializer": null,
      "lstm_weights_initializer": null,
      "_disable_preprocessor_api": false,
      "attention_memory_training": 50,
      "fcnet_weights_initializer": null,
      "_disable_action_flattening": false,
      "_use_default_native_models": -1,
      "attention_memory_inference": 50,
      "lstm_use_prev_action_reward": -1,
      "post_fcnet_bias_initializer": null,
      "attention_init_gru_gate_bias": 2.0,
      "attention_use_n_prev_actions": 0,
      "attention_use_n_prev_rewards": 0,
      "conv_bias_initializer_config": null,
      "lstm_bias_initializer_config": null,
      "fcnet_bias_initializer_config": null,
      "conv_kernel_initializer_config": null,
      "post_fcnet_weights_initializer": null,
      "attention_num_transformer_units": 1,
      "attention_position_wise_mlp_dim": 32,
      "conv_transpose_bias_initializer": null,
      "lstm_weights_initializer_config": null,
      "fcnet_weights_initializer_config": null,
      "conv_transpose_kernel_initializer": null,
      "post_fcnet_bias_initializer_config": null,
      "post_fcnet_weights_initializer_config": null,
      "conv_transpose_bias_initializer_config": null,
      "conv_transpose_kernel_initializer_config": null
    },
    "lambda": 0.9,
    "output": null,
    "explore": true,
    "use_gae": true,
    "kl_coeff": 0.2,
    "num_gpus": 0.0,
    "policies": {
      "default_policy": [
        null,
        null,
        null,
        null
      ]
    },
    "_is_atari": null,
    "callbacks": "ray.rllib.algorithms.callbacks.DefaultCallbacks",
    "framework": "torch",
    "grad_clip": null,
    "kl_target": 0.01,
    "log_level": "ERROR",
    "optimizer": {},
    "_fake_gpus": false,
    "batch_mode": "truncate_episodes",
    "clip_param": 0.4,
    "env_config": {},
    "render_env": false,
    "use_critic": true,
    "worker_cls": -1,
    "env_task_fn": null,
    "lr_schedule": null,
    "use_kl_loss": true,
    "action_space": null,
    "clip_actions": true,
    "clip_rewards": null,
    "fake_sampler": false,
    "grad_clip_by": "global_norm",
    "input_config": {},
    "num_learners": 0,
    "num_sgd_iter": 10,
    "eager_tracing": true,
    "entropy_coeff": 0.1,
    "in_evaluation": false,
    "local_gpu_idx": 0,
    "log_sys_usage": true,
    "logger_config": null,
    "output_config": {},
    "vf_clip_param": 10.0,
    "vf_loss_coeff": 0.25,
    "_learner_class": null,
    "count_steps_by": "env_steps",
    "env_runner_cls": null,
    "logger_creator": null,
    "observation_fn": null,
    "_rl_module_spec": null,
    "action_mask_key": "action_mask",
    "num_env_runners": 11,
    "tf_session_args": {
      "gpu_options": {
        "allow_growth": true
      },
      "device_count": {
        "CPU": 1
      },
      "allow_soft_placement": true,
      "log_device_placement": false,
      "inter_op_parallelism_threads": 2,
      "intra_op_parallelism_threads": 2
    },
    "vf_share_layers": -1,
    "offline_sampling": false,
    "policy_map_cache": -1,
    "prelearner_class": null,
    "sample_collector": "ray.rllib.evaluation.collectors.simple_list_collector.SimpleListCollector",
    "sample_timeout_s": 60.0,
    "simple_optimizer": -1,
    "train_batch_size": 512,
    "enable_connectors": true,
    "evaluation_config": null,
    "input_read_method": "read_parquet",
    "input_read_schema": {},
    "normalize_actions": true,
    "observation_space": null,
    "policies_to_train": null,
    "policy_mapping_fn": "ray.rllib.algorithms.algorithm_config.DEFAULT_POLICY_MAPPING_FN",
    "preprocessor_pref": "deepmind",
    "shuffle_sequences": true,
    "_learner_connector": null,
    "_model_config_dict": {},
    "eager_max_retraces": 20,
    "exploration_config": {
      "type": "StochasticSampling"
    },
    "map_batches_kwargs": {},
    "observation_filter": "NoFilter",
    "placement_strategy": "PACK",
    "postprocess_inputs": false,
    "remote_worker_envs": false,
    "sgd_minibatch_size": 64,
    "evaluation_duration": 10,
    "evaluation_interval": null,
    "iter_batches_kwargs": {},
    "learner_config_dict": {},
    "policy_map_capacity": 100,
    "shuffle_buffer_size": 0,
    "synchronize_filters": -1,
    "create_env_on_driver": false,
    "custom_eval_function": null,
    "disable_env_checking": -1,
    "num_cpus_per_learner": 1,
    "num_gpus_per_learner": 0,
    "output_max_file_size": 67108864,
    "torch_compile_worker": false,
    "_enable_rl_module_api": -1,
    "_per_module_overrides": {},
    "compress_observations": false,
    "local_tf_session_args": {
      "inter_op_parallelism_threads": 8,
      "intra_op_parallelism_threads": 8
    },
    "torch_compile_learner": false,
    "auto_wrap_old_gym_envs": -1,
    "entropy_coeff_schedule": null,
    "replay_sequence_length": null,
    "enable_async_evaluation": -1,
    "enable_tf1_exec_eagerly": false,
    "num_cpus_per_env_runner": 1,
    "num_envs_per_env_runner": 1,
    "num_gpus_per_env_runner": 0,
    "output_compress_columns": [
      "obs",
      "new_obs"
    ],
    "rollout_fragment_length": "auto",
    "use_worker_filter_stats": true,
    "_env_to_module_connector": null,
    "_module_to_env_connector": null,
    "episode_lookback_horizon": 1,
    "evaluation_duration_unit": "episodes",
    "input_read_method_kwargs": {},
    "min_time_s_per_iteration": null,
    "remote_env_batch_wait_ms": 0,
    "_disable_preprocessor_api": false,
    "export_native_model_files": false,
    "num_cpus_for_main_process": 1,
    "_disable_action_flattening": false,
    "evaluation_num_env_runners": 0,
    "ignore_env_runner_failures": false,
    "ope_split_batch_by_episode": true,
    "update_worker_filter_stats": true,
    "_disable_execution_plan_api": -1,
    "actions_in_input_normalized": false,
    "evaluation_sample_timeout_s": 120.0,
    "max_num_env_runner_restarts": 1000,
    "mini_batch_size_per_learner": null,
    "policy_states_are_swappable": false,
    "recreate_failed_env_runners": false,
    "sampler_perf_stats_ema_coef": null,
    "enable_rl_module_and_learner": false,
    "env_runner_restore_timeout_s": 1800,
    "train_batch_size_per_learner": null,
    "dataset_num_iters_per_learner": null,
    "off_policy_estimation_methods": {},
    "_run_training_always_in_thread": false,
    "prelearner_module_synch_period": 10,
    "custom_resources_per_env_runner": {},
    "evaluation_parallel_to_training": false,
    "keep_per_episode_custom_metrics": false,
    "restart_failed_sub_environments": false,
    "always_attach_evaluation_results": -1,
    "custom_async_evaluation_function": -1,
    "extra_python_environs_for_driver": {},
    "extra_python_environs_for_worker": {},
    "torch_compile_worker_dynamo_mode": null,
    "_dont_auto_sync_env_runner_states": false,
    "env_runner_health_probe_timeout_s": 30,
    "min_train_timesteps_per_iteration": 0,
    "torch_compile_learner_dynamo_mode": null,
    "checkpoint_trainable_policies_only": false,
    "enable_env_runner_and_connector_v2": false,
    "metrics_num_episodes_for_smoothing": 100,
    "min_sample_timesteps_per_iteration": 0,
    "delay_between_env_runner_restarts_s": 60.0,
    "torch_compile_worker_dynamo_backend": "onnxrt",
    "metrics_episode_collection_timeout_s": 60.0,
    "torch_compile_learner_dynamo_backend": "inductor",
    "_tf_policy_handles_more_than_one_loss": false,
    "algorithm_config_overrides_per_module": {},
    "torch_compile_learner_what_to_compile": "forward_train",
    "validate_env_runners_after_construction": true,
    "_disable_initialize_loss_from_dummy_batch": false,
    "max_requests_in_flight_per_sampler_worker": 2,
    "sync_filters_on_rollout_workers_timeout_s": 10.0,
    "_AlgorithmConfig__prior_exploration_config": null,
    "_evaluation_parallel_to_training_wo_thread": false,
    "add_default_connectors_to_learner_pipeline": true,
    "evaluation_force_reset_envs_before_iteration": true,
    "num_consecutive_env_runner_failures_tolerance": 100,
    "add_default_connectors_to_env_to_module_pipeline": true,
    "add_default_connectors_to_module_to_env_pipeline": true
  },
  "storage_path": "/tmp/ray_training_rtpnxl6x",
  "checkpoint_freq": 10,
  "checkpoint_at_end": true
}

{}

{
  "stop": {
    "timesteps_total": 5000000,
    "episode_len_mean": 300,
    "training_iteration": 1000000,
    "episode_reward_mean": 50
  },
  "config": {
    "lr": 2e-05,
    "env": "knights_archers_zombies_v10",
    "seed": null,
    "gamma": 0.99,
    "input": "sampler",
    "model": {
      "dim": 84,
      "use_lstm": false,
      "grayscale": false,
      "zero_mean": true,
      "framestack": true,
      "_time_major": false,
      "max_seq_len": 20,
      "conv_filters": null,
      "custom_model": null,
      "free_log_std": false,
      "attention_dim": 64,
      "fcnet_hiddens": [
        256,
        256
      ],
      "use_attention": false,
      "lstm_cell_size": 256,
      "conv_activation": "relu",
      "no_final_linear": false,
      "vf_share_layers": false,
      "fcnet_activation": "tanh",
      "attention_head_dim": 32,
      "custom_action_dist": null,
      "encoder_latent_dim": null,
      "post_fcnet_hiddens": [],
      "always_check_shapes": false,
      "attention_num_heads": 1,
      "custom_model_config": {},
      "custom_preprocessor": null,
      "lstm_use_prev_action": false,
      "lstm_use_prev_reward": false,
      "conv_bias_initializer": null,
      "lstm_bias_initializer": null,
      "post_fcnet_activation": "relu",
      "fcnet_bias_initializer": null,
      "conv_kernel_initializer": null,
      "lstm_weights_initializer": null,
      "_disable_preprocessor_api": false,
      "attention_memory_training": 50,
      "fcnet_weights_initializer": null,
      "_disable_action_flattening": false,
      "_use_default_native_models": -1,
      "attention_memory_inference": 50,
      "lstm_use_prev_action_reward": -1,
      "post_fcnet_bias_initializer": null,
      "attention_init_gru_gate_bias": 2.0,
      "attention_use_n_prev_actions": 0,
      "attention_use_n_prev_rewards": 0,
      "conv_bias_initializer_config": null,
      "lstm_bias_initializer_config": null,
      "fcnet_bias_initializer_config": null,
      "conv_kernel_initializer_config": null,
      "post_fcnet_weights_initializer": null,
      "attention_num_transformer_units": 1,
      "attention_position_wise_mlp_dim": 32,
      "conv_transpose_bias_initializer": null,
      "lstm_weights_initializer_config": null,
      "fcnet_weights_initializer_config": null,
      "conv_transpose_kernel_initializer": null,
      "post_fcnet_bias_initializer_config": null,
      "post_fcnet_weights_initializer_config": null,
      "conv_transpose_bias_initializer_config": null,
      "conv_transpose_kernel_initializer_config": null
    },
    "lambda": 0.9,
    "output": null,
    "explore": true,
    "use_gae": true,
    "kl_coeff": 0.2,
    "num_gpus": 0.0,
    "policies": {
      "default_policy": [
        null,
        null,
        null,
        null
      ]
    },
    "_is_atari": null,
    "callbacks": "ray.rllib.algorithms.callbacks.DefaultCallbacks",
    "framework": "torch",
    "grad_clip": null,
    "kl_target": 0.01,
    "log_level": "ERROR",
    "optimizer": {},
    "_fake_gpus": false,
    "batch_mode": "truncate_episodes",
    "clip_param": 0.4,
    "env_config": {},
    "render_env": false,
    "use_critic": true,
    "worker_cls": -1,
    "env_task_fn": null,
    "lr_schedule": null,
    "use_kl_loss": true,
    "action_space": null,
    "clip_actions": true,
    "clip_rewards": null,
    "fake_sampler": false,
    "grad_clip_by": "global_norm",
    "input_config": {},
    "num_learners": 0,
    "num_sgd_iter": 10,
    "eager_tracing": true,
    "entropy_coeff": 0.1,
    "in_evaluation": false,
    "local_gpu_idx": 0,
    "log_sys_usage": true,
    "logger_config": null,
    "output_config": {},
    "vf_clip_param": 10.0,
    "vf_loss_coeff": 0.25,
    "_learner_class": null,
    "count_steps_by": "env_steps",
    "env_runner_cls": null,
    "logger_creator": null,
    "observation_fn": null,
    "_rl_module_spec": null,
    "action_mask_key": "action_mask",
    "num_env_runners": 11,
    "tf_session_args": {
      "gpu_options": {
        "allow_growth": true
      },
      "device_count": {
        "CPU": 1
      },
      "allow_soft_placement": true,
      "log_device_placement": false,
      "inter_op_parallelism_threads": 2,
      "intra_op_parallelism_threads": 2
    },
    "vf_share_layers": -1,
    "offline_sampling": false,
    "policy_map_cache": -1,
    "prelearner_class": null,
    "sample_collector": "ray.rllib.evaluation.collectors.simple_list_collector.SimpleListCollector",
    "sample_timeout_s": 60.0,
    "simple_optimizer": -1,
    "train_batch_size": 512,
    "enable_connectors": true,
    "evaluation_config": null,
    "input_read_method": "read_parquet",
    "input_read_schema": {},
    "normalize_actions": true,
    "observation_space": null,
    "policies_to_train": null,
    "policy_mapping_fn": "ray.rllib.algorithms.algorithm_config.DEFAULT_POLICY_MAPPING_FN",
    "preprocessor_pref": "deepmind",
    "shuffle_sequences": true,
    "_learner_connector": null,
    "_model_config_dict": {},
    "eager_max_retraces": 20,
    "exploration_config": {
      "type": "StochasticSampling"
    },
    "map_batches_kwargs": {},
    "observation_filter": "NoFilter",
    "placement_strategy": "PACK",
    "postprocess_inputs": false,
    "remote_worker_envs": false,
    "sgd_minibatch_size": 64,
    "evaluation_duration": 10,
    "evaluation_interval": null,
    "iter_batches_kwargs": {},
    "learner_config_dict": {},
    "policy_map_capacity": 100,
    "shuffle_buffer_size": 0,
    "synchronize_filters": -1,
    "create_env_on_driver": false,
    "custom_eval_function": null,
    "disable_env_checking": -1,
    "num_cpus_per_learner": 1,
    "num_gpus_per_learner": 0,
    "output_max_file_size": 67108864,
    "torch_compile_worker": false,
    "_enable_rl_module_api": -1,
    "_per_module_overrides": {},
    "compress_observations": false,
    "local_tf_session_args": {
      "inter_op_parallelism_threads": 8,
      "intra_op_parallelism_threads": 8
    },
    "torch_compile_learner": false,
    "auto_wrap_old_gym_envs": -1,
    "entropy_coeff_schedule": null,
    "replay_sequence_length": null,
    "enable_async_evaluation": -1,
    "enable_tf1_exec_eagerly": false,
    "num_cpus_per_env_runner": 1,
    "num_envs_per_env_runner": 1,
    "num_gpus_per_env_runner": 0,
    "output_compress_columns": [
      "obs",
      "new_obs"
    ],
    "rollout_fragment_length": "auto",
    "use_worker_filter_stats": true,
    "_env_to_module_connector": null,
    "_module_to_env_connector": null,
    "episode_lookback_horizon": 1,
    "evaluation_duration_unit": "episodes",
    "input_read_method_kwargs": {},
    "min_time_s_per_iteration": null,
    "remote_env_batch_wait_ms": 0,
    "_disable_preprocessor_api": false,
    "export_native_model_files": false,
    "num_cpus_for_main_process": 1,
    "_disable_action_flattening": false,
    "evaluation_num_env_runners": 0,
    "ignore_env_runner_failures": false,
    "ope_split_batch_by_episode": true,
    "update_worker_filter_stats": true,
    "_disable_execution_plan_api": -1,
    "actions_in_input_normalized": false,
    "evaluation_sample_timeout_s": 120.0,
    "max_num_env_runner_restarts": 1000,
    "mini_batch_size_per_learner": null,
    "policy_states_are_swappable": false,
    "recreate_failed_env_runners": false,
    "sampler_perf_stats_ema_coef": null,
    "enable_rl_module_and_learner": false,
    "env_runner_restore_timeout_s": 1800,
    "train_batch_size_per_learner": null,
    "dataset_num_iters_per_learner": null,
    "off_policy_estimation_methods": {},
    "_run_training_always_in_thread": false,
    "prelearner_module_synch_period": 10,
    "custom_resources_per_env_runner": {},
    "evaluation_parallel_to_training": false,
    "keep_per_episode_custom_metrics": false,
    "restart_failed_sub_environments": false,
    "always_attach_evaluation_results": -1,
    "custom_async_evaluation_function": -1,
    "extra_python_environs_for_driver": {},
    "extra_python_environs_for_worker": {},
    "torch_compile_worker_dynamo_mode": null,
    "_dont_auto_sync_env_runner_states": false,
    "env_runner_health_probe_timeout_s": 30,
    "min_train_timesteps_per_iteration": 0,
    "torch_compile_learner_dynamo_mode": null,
    "checkpoint_trainable_policies_only": false,
    "enable_env_runner_and_connector_v2": false,
    "metrics_num_episodes_for_smoothing": 100,
    "min_sample_timesteps_per_iteration": 0,
    "delay_between_env_runner_restarts_s": 60.0,
    "torch_compile_worker_dynamo_backend": "onnxrt",
    "metrics_episode_collection_timeout_s": 60.0,
    "torch_compile_learner_dynamo_backend": "inductor",
    "_tf_policy_handles_more_than_one_loss": false,
    "algorithm_config_overrides_per_module": {},
    "torch_compile_learner_what_to_compile": "forward_train",
    "validate_env_runners_after_construction": true,
    "_disable_initialize_loss_from_dummy_batch": false,
    "max_requests_in_flight_per_sampler_worker": 2,
    "sync_filters_on_rollout_workers_timeout_s": 10.0,
    "_AlgorithmConfig__prior_exploration_config": null,
    "_evaluation_parallel_to_training_wo_thread": false,
    "add_default_connectors_to_learner_pipeline": true,
    "evaluation_force_reset_envs_before_iteration": true,
    "num_consecutive_env_runner_failures_tolerance": 100,
    "add_default_connectors_to_env_to_module_pipeline": true,
    "add_default_connectors_to_module_to_env_pipeline": true
  },
  "storage_path": "/tmp/ray_training_rtpnxl6x",
  "checkpoint_freq": 10,
  "checkpoint_at_end": true
}

{}