{ "condition": "clean_kp", "config": { "action_dim": 7, "context_gain": 1.0, "hidden_layers": 4, "input_dim": 3964, "learning_rate": 0.03, "mission_dim": 23, "momentum": 0.9, "reciprocal_learning_rate": 0.03, "weight_decay": 0.0001, "width": 256 }, "data": { "color_cardinality": 6, "elapsed_seconds": 81.82925486564636, "env_id": "BabyAI-PickupLoc-v0", "minigrid_version": "3.1.0", "object_cardinality": 11, "protocol": "babyai_shared_feedback_b0", "rollout_episodes": 500, "rollout_seed_start": 200000, "state_cardinality": 3, "train_episodes": 20000, "train_steps": 121467, "validation_episodes": 2000, "validation_steps": 12347, "vocabulary": [ "", "a", "ball", "behind", "blue", "box", "front", "green", "grey", "in", "key", "left", "of", "on", "pick", "purple", "red", "right", "the", "up", "yellow", "you", "your" ] }, "epoch_history": [ { "epoch": 1, "train_loss": 0.6793887242517973 }, { "epoch": 2, "train_loss": 0.5070055065029546 }, { "epoch": 3, "train_loss": 0.4780606180743167 }, { "epoch": 4, "train_loss": 0.46464681211270786 }, { "epoch": 5, "train_loss": 0.45651326932405173 }, { "epoch": 6, "train_loss": 0.4414772361203244 }, { "epoch": 7, "train_loss": 0.4295881901916705 }, { "epoch": 8, "train_loss": 0.4249380529554267 }, { "epoch": 9, "train_loss": 0.4178563620542225 }, { "epoch": 10, "train_loss": 0.40975865175849513 }, { "epoch": 11, "train_loss": 0.40546370631770084 }, { "epoch": 12, "train_loss": 0.3982252176811821 }, { "epoch": 13, "train_loss": 0.3858343799490678 }, { "epoch": 14, "train_loss": 0.379007538933503 }, { "epoch": 15, "train_loss": 0.3656208556576779 }, { "epoch": 16, "train_loss": 0.35379995308424295 }, { "epoch": 17, "train_loss": 0.33681938353337737 }, { "epoch": 18, "train_loss": 0.32208908896697197 }, { "epoch": 19, "train_loss": 0.30617407221543164 }, { "epoch": 20, "train_loss": 0.2926546273419732 }, { "epoch": 21, "train_loss": 0.28012341188757045 }, { "epoch": 22, "train_loss": 0.27170656586948194 }, { "epoch": 23, "train_loss": 0.2618562983211718 }, { "epoch": 24, "train_loss": 0.25525567559819473 }, { "epoch": 25, "train_loss": 0.24340780433855558 } ], "epochs_completed": 25, "finite": true, "first_nonfinite_epoch": null, "mission_lesion_rollout": { "episodes": 500, "mean_length": 44.094, "mean_return": 0.32912812499999994, "success": 0.492 }, "mission_lesion_validation": { "accuracy": 0.7498177694986636, "loss": 0.6379798919437983 }, "predictor": [], "provenance": { "cuda_device_name": "NVIDIA GeForce GTX 1080", "cuda_peak_allocated_bytes": 130979840, "cuda_visible_devices": "1", "device": "cuda:0", "git_commit": "a30c297eb8abf6f15f97dd4f35a2f530c9728100", "git_dirty_tracked": false, "minigrid_version": "3.1.0", "torch_version": "2.3.1+cu118" }, "rollout": { "episodes": 500, "mean_length": 31.056, "mean_return": 0.524875, "success": 0.616 }, "stage": "babyai_pickup_p1", "training": { "batch_size": 256, "history_steps": 4, "model_seed": 4101, "neutral_examples_per_epoch": 0, "parameter_count_including_reciprocal_context_predictor": 1438215, "shuffle_seed": 4101, "total_wall_seconds_including_rollouts": 112.63866543769836, "training_wall_seconds": 98.61020755767822 }, "validation": { "accuracy": 0.8578602089576415, "loss": 0.31587319402276837 } }