comprehensive_results_binary.json

{
  "val": {
    "loss": 0.2408331334590912,
    "accuracy": 0.9273356401384083,
    "precision": 0.9481481481481482,
    "recall": 0.9014084507042254,
    "f1": 0.924187725631769,
    "macro_f1": 0.927210141885652,
    "roc_auc": 0.9746095621347131,
    "non_gas_precision": 0.9090909090909091,
    "non_gas_recall": 0.9523809523809523,
    "non_gas_f1": 0.9302325581395349,
    "runtime": 13.2248,
    "samples_per_second": 21.853,
    "steps_per_second": 1.437,
    "confusion_matrix": [
      [
        140,
        7
      ],
      [
        14,
        128
      ]
    ]
  },
  "test_id": {
    "loss": 0.33420857787132263,
    "accuracy": 0.8927335640138409,
    "precision": 0.8992805755395683,
    "recall": 0.8802816901408451,
    "f1": 0.8896797153024911,
    "macro_f1": 0.8926513054626934,
    "roc_auc": 0.951806074542493,
    "non_gas_precision": 0.8866666666666667,
    "non_gas_recall": 0.9047619047619048,
    "non_gas_f1": 0.8956228956228957,
    "runtime": 13.0723,
    "samples_per_second": 22.108,
    "steps_per_second": 1.453,
    "confusion_matrix": [
      [
        133,
        14
      ],
      [
        17,
        125
      ]
    ]
  },
  "test_ood": {
    "loss": 0.5717462301254272,
    "accuracy": 0.8170289855072463,
    "precision": 0.9970414201183432,
    "recall": 0.7711670480549199,
    "f1": 0.8696774193548387,
    "macro_f1": 0.781343268947936,
    "roc_auc": 0.9866480947169436,
    "non_gas_precision": 0.5327102803738317,
    "non_gas_recall": 0.991304347826087,
    "non_gas_f1": 0.6930091185410334,
    "runtime": 13.3686,
    "samples_per_second": 41.291,
    "steps_per_second": 2.618,
    "confusion_matrix": [
      [
        114,
        1
      ],
      [
        100,
        337
      ]
    ]
  },
  "delta_f1": 0.02000229594765246,
  "delta_f1_interpretation": "\u2705 Robust generalization (minimal domain sensitivity)"
}