metadata/checkpoint_summary.json

{
  "actor_critic_state_key_count": 13,
  "actor_critic_state_shapes": {
    "policy_net.latent_layer.bias": [
      128
    ],
    "policy_net.latent_layer.weight": [
      128,
      576
    ],
    "policy_net.log_std": [
      6
    ],
    "policy_net.trunk.0.bias": [
      128
    ],
    "policy_net.trunk.0.weight": [
      128,
      130
    ],
    "policy_net.trunk.2.bias": [
      6
    ],
    "policy_net.trunk.2.weight": [
      6,
      128
    ],
    "value_net.latent_layer.bias": [
      128
    ],
    "value_net.latent_layer.weight": [
      128,
      576
    ],
    "value_net.trunk.0.bias": [
      128
    ],
    "value_net.trunk.0.weight": [
      128,
      130
    ],
    "value_net.trunk.2.bias": [
      1
    ],
    "value_net.trunk.2.weight": [
      1,
      128
    ]
  },
  "best_iteration": 465,
  "best_metric_mode": "max",
  "best_metric_name": "validation/outcome/success_lift_vs_dataset",
  "best_metric_value": 0.109375,
  "completed_iterations": 466,
  "experiment_name": "seed8_rwd-grp-b_stb-rwd-5x_128-epi_paper-spec_latefus",
  "object_split": {
    "split_seed": 7,
    "test_ids": [
      75,
      76,
      77,
      79,
      80,
      81,
      83,
      84,
      86
    ],
    "train_ids": [
      0,
      1,
      2,
      3,
      4,
      5,
      6,
      7,
      8,
      9,
      10,
      11,
      12,
      13,
      14,
      15,
      16,
      17,
      18,
      19,
      20,
      21,
      22,
      23,
      24,
      25,
      26,
      27,
      28,
      29,
      30,
      31,
      32,
      33,
      34,
      35,
      36,
      37,
      38,
      39,
      40,
      41,
      42,
      43,
      44,
      45,
      46,
      47,
      48,
      49,
      50,
      51,
      52,
      53,
      54,
      55,
      56,
      57,
      58,
      59,
      60,
      61,
      62,
      63,
      64,
      65,
      66,
      67,
      68,
      69,
      70,
      71,
      72,
      73,
      74
    ],
    "val_ids": [
      78,
      82,
      85,
      87
    ]
  },
  "observed_environment": {
    "cuda_available_at_packaging": false,
    "python": "3.10.12",
    "torch": "2.4.1+cu124",
    "torch_cuda": "12.4"
  },
  "omitted_for_public_eval": [
    "optimizer",
    "history"
  ],
  "public_checkpoint_keys": [
    "actor_critic",
    "best_iteration",
    "best_metric_mode",
    "best_metric_name",
    "best_metric_value",
    "calibrator",
    "completed_iterations",
    "experiment_cfg",
    "object_split",
    "omitted_keys",
    "release_format",
    "source_checkpoint_path",
    "source_experiment_dir"
  ],
  "release_checkpoint": "checkpoints/full_sga_gsn_seed8_best.pt",
  "seed": 8,
  "source_checkpoint_bytes": 4527762,
  "source_checkpoint_keys": [
    "actor_critic",
    "best_iteration",
    "best_metric_mode",
    "best_metric_name",
    "best_metric_value",
    "calibrator",
    "completed_iterations",
    "experiment_cfg",
    "history",
    "object_split",
    "optimizer"
  ],
  "source_checkpoint_path": "/rl-grasp-refine/outputs/exp_debug/seed8_rwd-grp-b_stb-rwd-5x_128-epi_paper-spec_latefus/checkpoints/best.pt",
  "source_experiment_dir": "/rl-grasp-refine/outputs/exp_debug/seed8_rwd-grp-b_stb-rwd-5x_128-epi_paper-spec_latefus"
}