-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
Showing
10 changed files
with
2,087 additions
and
307 deletions.
There are no files selected for viewing
This file contains 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
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,354 @@ | ||
| __module__: src.trainer.config | ||
| __name__: AlphaZeroTrainerConfig | ||
| data: | ||
| collector_cfg: | ||
| __module__: src.trainer.config | ||
| __name__: CollectorConfig | ||
| data: | ||
| buffer_size: 100000 | ||
| log_every_sec: 300 | ||
| quick_start_buffer_path: null | ||
| start_wait_n_samples: 100000 | ||
| compile_mode: reduce-overhead | ||
| compile_model: false | ||
| data_qsize: 10 | ||
| distributor_out_qsize: 10 | ||
| evaluator_cfg: | ||
| __module__: src.trainer.config | ||
| __name__: EvaluatorConfig | ||
| data: | ||
| enemy_cfgs: | ||
| - __module__: src.agent.one_shot | ||
| __name__: RandomAgentConfig | ||
| data: | ||
| name: RandomAgent | ||
| enemy_iterations: 100 | ||
| eval_rate_sec: 20 | ||
| num_episodes: 100 | ||
| prevent_draw: true | ||
| save_checkpoints: false | ||
| temperature: 1 | ||
| game_cfg: | ||
| __module__: src.game.overcooked.config | ||
| __name__: CrampedRoomOvercookedConfig | ||
| data: | ||
| board: | ||
| - - 1 | ||
| - 1 | ||
| - 4 | ||
| - 1 | ||
| - 1 | ||
| - - 3 | ||
| - 0 | ||
| - 0 | ||
| - 0 | ||
| - 3 | ||
| - - 1 | ||
| - 0 | ||
| - 0 | ||
| - 0 | ||
| - 1 | ||
| - - 1 | ||
| - 2 | ||
| - 1 | ||
| - 5 | ||
| - 1 | ||
| cooking_time: 20 | ||
| flat_obs: false | ||
| h: 4 | ||
| horizon: 400 | ||
| num_actions: 6 | ||
| num_players: 2 | ||
| reward_cfg: | ||
| __module__: src.game.overcooked.config | ||
| __name__: OvercookedRewardConfig | ||
| data: | ||
| dish_pickup: 3 | ||
| placement_in_pot: 3 | ||
| soup_delivery: 20 | ||
| soup_pickup: 5 | ||
| start_cooking: 3 | ||
| reward_scaling_factor: 0.5 | ||
| single_temperature_input: true | ||
| start_pos: | ||
| __module__: src.misc.serialization | ||
| __name__: TupleWrapper | ||
| data: | ||
| data: | ||
| - __module__: src.misc.serialization | ||
| __name__: TupleWrapper | ||
| data: | ||
| data: | ||
| - 1 | ||
| - 2 | ||
| - 0 | ||
| - 0 | ||
| - __module__: src.misc.serialization | ||
| __name__: TupleWrapper | ||
| data: | ||
| data: | ||
| - 3 | ||
| - 1 | ||
| - 0 | ||
| - 0 | ||
| temperature_input: true | ||
| w: 5 | ||
| inf_cfg: | ||
| __module__: src.trainer.config | ||
| __name__: InferenceServerConfig | ||
| data: | ||
| statistics_every_sec: 60 | ||
| use_gpu: true | ||
| info_qsize: 100 | ||
| init_new_network_params: false | ||
| logger_cfg: | ||
| __module__: src.trainer.config | ||
| __name__: LoggerConfig | ||
| data: | ||
| buffer_gen: false | ||
| id: 0 | ||
| name: oc_proxy | ||
| project_name: overcooked_cramped | ||
| updater_bucket_size: 100 | ||
| wandb_mode: online | ||
| worker_episode_bucket_size: 25 | ||
| max_batch_size: 3000 | ||
| max_cpu_evaluator: 1 | ||
| max_cpu_inference_server: 2 | ||
| max_cpu_log_dist_save_collect: 1 | ||
| max_cpu_updater: 2 | ||
| max_cpu_worker: 11 | ||
| max_eval_per_worker: 6000 | ||
| merge_inference_update_gpu: false | ||
| net_cfg: | ||
| __module__: src.network.resnet | ||
| __name__: OvercookedResNetConfig5x5 | ||
| data: | ||
| activation_type: | ||
| __module__: src.network.utils | ||
| __name__: ActivationType | ||
| value: LEAKY_RELU | ||
| eq_type: | ||
| __module__: src.network.vision_net | ||
| __name__: EquivarianceType | ||
| value: NONE | ||
| game_cfg: null | ||
| layer_specs: | ||
| - - 32 | ||
| - 3 | ||
| - 3 | ||
| - 1 | ||
| - 1 | ||
| - - 64 | ||
| - 2 | ||
| - 3 | ||
| - 1 | ||
| - 1 | ||
| - - 128 | ||
| - 1 | ||
| - 3 | ||
| - 1 | ||
| - 1 | ||
| - - 256 | ||
| - 1 | ||
| - 3 | ||
| - 0 | ||
| - 1 | ||
| lff_feature_expansion: 40 | ||
| lff_features: false | ||
| norm_type: | ||
| __module__: src.network.utils | ||
| __name__: NormalizationType | ||
| value: GROUP_NORM | ||
| policy_head_cfg: | ||
| __module__: src.network.fcn | ||
| __name__: WideHeadConfig | ||
| data: | ||
| activation_type: | ||
| __module__: src.network.utils | ||
| __name__: ActivationType | ||
| value: LEAKY_RELU | ||
| dropout_p: 0.2 | ||
| final_activation: | ||
| __module__: src.network.utils | ||
| __name__: ActivationType | ||
| value: NONE | ||
| hidden_size: 256 | ||
| normalization_type: | ||
| __module__: src.network.utils | ||
| __name__: NormalizationType | ||
| value: GROUP_NORM | ||
| num_layers: 1 | ||
| predict_policy: true | ||
| value_head_cfg: | ||
| __module__: src.network.fcn | ||
| __name__: WideHeadConfig | ||
| data: | ||
| activation_type: | ||
| __module__: src.network.utils | ||
| __name__: ActivationType | ||
| value: LEAKY_RELU | ||
| dropout_p: 0.2 | ||
| final_activation: | ||
| __module__: src.network.utils | ||
| __name__: ActivationType | ||
| value: NONE | ||
| hidden_size: 256 | ||
| normalization_type: | ||
| __module__: src.network.utils | ||
| __name__: NormalizationType | ||
| value: GROUP_NORM | ||
| num_layers: 1 | ||
| num_inference_server: 1 | ||
| num_worker: 30 | ||
| only_generate_buffer: false | ||
| prev_run_dir: null | ||
| prev_run_idx: null | ||
| proxy_net_path: null | ||
| restrict_cpu: true | ||
| save_state: false | ||
| save_state_after_seconds: 30 | ||
| saver_cfg: | ||
| __module__: src.trainer.config | ||
| __name__: SaverConfig | ||
| data: | ||
| save_all_checkpoints: false | ||
| save_interval_sec: 30 | ||
| single_sbr_temperature: true | ||
| temperature_input: true | ||
| updater_cfg: | ||
| __module__: src.trainer.config | ||
| __name__: UpdaterConfig | ||
| data: | ||
| gradient_max_norm: 1.0 | ||
| mse_policy_loss: true | ||
| optim_cfg: | ||
| __module__: src.supervised.optim | ||
| __name__: OptimizerConfig | ||
| data: | ||
| anneal_cfg: | ||
| __module__: src.supervised.annealer | ||
| __name__: TemperatureAnnealingConfig | ||
| data: | ||
| anneal_temps: | ||
| - 0.001 | ||
| - 1.0e-06 | ||
| anneal_types: | ||
| - __module__: src.supervised.annealer | ||
| __name__: AnnealingType | ||
| value: LINEAR | ||
| - __module__: src.supervised.annealer | ||
| __name__: AnnealingType | ||
| value: COSINE | ||
| cyclic: false | ||
| end_times_min: | ||
| - 30 | ||
| - 1400 | ||
| init_temp: 0 | ||
| sampling: false | ||
| beta1: 0.9 | ||
| beta2: 0.99 | ||
| fused: false | ||
| optim_type: | ||
| __module__: src.supervised.optim | ||
| __name__: OptimType | ||
| value: ADAM_W | ||
| weight_decay: 0.0001 | ||
| policy_loss_factor: 5 | ||
| updates_until_distribution: 5 | ||
| use_gpu: true | ||
| utility_loss: | ||
| __module__: src.game.values | ||
| __name__: UtilityNorm | ||
| value: FULL_COOP | ||
| utility_loss_factor: 1 | ||
| value_reg_loss_factor: 0 | ||
| updater_in_qsize: 100 | ||
| updater_out_qsize: 10 | ||
| validator_data_qsize: 100 | ||
| worker_cfg: | ||
| __module__: src.trainer.config | ||
| __name__: WorkerConfig | ||
| data: | ||
| anneal_cfgs: | ||
| - __module__: src.supervised.annealer | ||
| __name__: TemperatureAnnealingConfig | ||
| data: | ||
| anneal_temps: | ||
| - 10 | ||
| anneal_types: | ||
| - __module__: src.supervised.annealer | ||
| __name__: AnnealingType | ||
| value: COSINE | ||
| cyclic: true | ||
| end_times_min: | ||
| - 1 | ||
| init_temp: 0 | ||
| sampling: true | ||
| exploration_prob: 0 | ||
| max_game_length: 8 | ||
| max_random_start_steps: 0 | ||
| policy_eval_cfg: | ||
| __module__: src.trainer.policy_eval | ||
| __name__: PolicyEvalConfig | ||
| data: | ||
| eval_type: | ||
| __module__: src.trainer.policy_eval | ||
| __name__: PolicyEvalType | ||
| value: TD_0 | ||
| lambda_val: 0.5 | ||
| prevent_draw: false | ||
| quick_start: false | ||
| search_cfg: | ||
| __module__: src.search.config | ||
| __name__: FixedDepthConfig | ||
| data: | ||
| average_eval: false | ||
| backup_func_cfg: | ||
| __module__: src.search.config | ||
| __name__: LogitBackupConfig | ||
| data: | ||
| epsilon: 0 | ||
| hp_0: null | ||
| hp_1: null | ||
| init_random: true | ||
| init_temperatures: | ||
| - 15 | ||
| - 15 | ||
| num_iterations: 150 | ||
| sbr_mode: | ||
| __module__: src.equilibria.logit | ||
| __name__: SbrMode | ||
| value: NAGURNEY | ||
| use_cpp: true | ||
| discount: 0.93 | ||
| eval_func_cfg: | ||
| __module__: src.search.config | ||
| __name__: InferenceServerEvalConfig | ||
| data: | ||
| active_wait_time: 0.05 | ||
| init_temperatures: null | ||
| max_clip_value: 20 | ||
| min_clip_value: -.inf | ||
| policy_prediction: true | ||
| random_symmetry: false | ||
| single_temperature: true | ||
| temperature_input: true | ||
| utility_norm: | ||
| __module__: src.game.values | ||
| __name__: UtilityNorm | ||
| value: NONE | ||
| extract_func_cfg: | ||
| __module__: src.search.config | ||
| __name__: SpecialExtractConfig | ||
| data: | ||
| utility_norm: | ||
| __module__: src.game.values | ||
| __name__: UtilityNorm | ||
| value: FULL_COOP | ||
| search_iterations: 1 | ||
| temperature: 1 | ||
| use_symmetries: true | ||
| hydra: | ||
| run: | ||
| dir: ./outputs/${now:%Y-%m-%d}/${now:%H-%M-%S}_oc_proxy |
Oops, something went wrong.