{ "condition": "bp", "config": { "action_dim": 7, "context_gain": 1.0, "hidden_layers": 4, "input_dim": 7928, "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": 0.6321194964961002 }, { "epoch": 2, "train_loss": 0.4924550595408992 }, { "epoch": 3, "train_loss": 0.4738408164601577 }, { "epoch": 4, "train_loss": 0.44879562810847634 }, { "epoch": 5, "train_loss": 0.4285990457785757 }, { "epoch": 6, "train_loss": 0.41075110887226307 }, { "epoch": 7, "train_loss": 0.39140965926019766 }, { "epoch": 8, "train_loss": 0.37369739294052123 }, { "epoch": 9, "train_loss": 0.34568895095273067 }, { "epoch": 10, "train_loss": 0.3196536517456958 }, { "epoch": 11, "train_loss": 0.30388873991213344 }, { "epoch": 12, "train_loss": 0.28137334092667227 }, { "epoch": 13, "train_loss": 0.27270261155931574 }, { "epoch": 14, "train_loss": 0.2571609906773818 }, { "epoch": 15, "train_loss": 0.24464017152786255 }, { "epoch": 16, "train_loss": 0.2372008582792784 }, { "epoch": 17, "train_loss": 0.23206563582545833 }, { "epoch": 18, "train_loss": 0.22314524976830732 }, { "epoch": 19, "train_loss": 0.21708365371352747 }, { "epoch": 20, "train_loss": 0.20607339666077965 }, { "epoch": 21, "train_loss": 0.20372903077225937 }, { "epoch": 22, "train_loss": 0.19777495506562684 }, { "epoch": 23, "train_loss": 0.19298629737214038 }, { "epoch": 24, "train_loss": 0.18727699292333502 }, { "epoch": 25, "train_loss": 0.18384965098217915 } ], "epochs_completed": 25, "finite": true, "first_nonfinite_epoch": null, "mission_lesion_rollout": { "episodes": 500, "mean_length": 46.88, "mean_return": 0.27814999999999995, "success": 0.374 }, "mission_lesion_validation": { "accuracy": 0.7470640641451365, "loss": 0.7762952407400666 }, "predictor": [], "provenance": { "cuda_device_name": "NVIDIA GeForce GTX 1080", "cuda_peak_allocated_bytes": 247702528, "cuda_visible_devices": "2", "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": 28.428, "mean_return": 0.56743125, "success": 0.672 }, "stage": "babyai_pickup_p1", "training": { "batch_size": 256, "history_steps": 8, "model_seed": 4101, "neutral_examples_per_epoch": 0, "parameter_count_including_reciprocal_context_predictor": 2452999, "shuffle_seed": 4101, "total_wall_seconds_including_rollouts": 113.33854031562805, "training_wall_seconds": 97.784170627594 }, "validation": { "accuracy": 0.8668502470235685, "loss": 0.32782987784767786 } }