detailed_results_validation.json

{
  "split": "validation",
  "classification_report": "              precision    recall  f1-score   support\n\n           O     0.9736    0.9813    0.9774     17717\n       COMMA     0.6741    0.6105    0.6408       986\n        DARI     0.7895    0.7971    0.7933       414\n    QUESTION     0.8565    0.8685    0.8625       213\n EXCLAMATION     0.5909    0.5417    0.5652        48\n   SEMICOLON     0.7333    0.4400    0.5500        25\n       COLON     0.5614    0.4211    0.4812        76\n      HYPHEN     0.4286    0.1111    0.1765        27\n\n    accuracy                         0.9522     19506\n   macro avg     0.7010    0.5964    0.6309     19506\nweighted avg     0.9497    0.9522    0.9506     19506\n",
  "confusion_matrix": [
    [
      17385,
      246,
      47,
      17,
      3,
      1,
      17,
      1
    ],
    [
      342,
      602,
      28,
      3,
      2,
      2,
      4,
      3
    ],
    [
      40,
      28,
      330,
      4,
      10,
      1,
      1,
      0
    ],
    [
      18,
      1,
      6,
      185,
      3,
      0,
      0,
      0
    ],
    [
      5,
      3,
      7,
      7,
      26,
      0,
      0,
      0
    ],
    [
      12,
      2,
      0,
      0,
      0,
      11,
      0,
      0
    ],
    [
      34,
      10,
      0,
      0,
      0,
      0,
      32,
      0
    ],
    [
      20,
      1,
      0,
      0,
      0,
      0,
      3,
      3
    ]
  ],
  "per_class_metrics": [
    {
      "class": "O",
      "support": 17717,
      "accuracy": 0.9812609358243495,
      "correct": 17385
    },
    {
      "class": "COMMA",
      "support": 986,
      "accuracy": 0.6105476673427992,
      "correct": 602
    },
    {
      "class": "DARI",
      "support": 414,
      "accuracy": 0.7971014492753623,
      "correct": 330
    },
    {
      "class": "QUESTION",
      "support": 213,
      "accuracy": 0.8685446009389671,
      "correct": 185
    },
    {
      "class": "EXCLAMATION",
      "support": 48,
      "accuracy": 0.5416666666666666,
      "correct": 26
    },
    {
      "class": "SEMICOLON",
      "support": 25,
      "accuracy": 0.44,
      "correct": 11
    },
    {
      "class": "COLON",
      "support": 76,
      "accuracy": 0.42105263157894735,
      "correct": 32
    },
    {
      "class": "HYPHEN",
      "support": 27,
      "accuracy": 0.1111111111111111,
      "correct": 3
    }
  ],
  "bleu_rouge_scores": {
    "bleu": 90.20234005231055,
    "rouge1_precision": 0.0,
    "rouge1_recall": 0.0,
    "rouge1_f": 0.0,
    "rouge2_precision": 0.0,
    "rouge2_recall": 0.0,
    "rouge2_f": 0.0,
    "rougeL_precision": 0.0,
    "rougeL_recall": 0.0,
    "rougeL_f": 0.0
  },
  "overall_metrics": {
    "test_loss": 0.023327844217419624,
    "test_precision": 0.9496856615053557,
    "test_recall": 0.9522198297959602,
    "test_f1": 0.950640869722443,
    "test_weighted_f1": 0.950640869722443,
    "test_punctuation_f1": 0.7575861893899899,
    "test_f1_comma": 0.6407663650878127,
    "test_f1_dari": 0.7932692307692307,
    "test_f1_question": 0.8624708624708625,
    "test_f1_exclamation": 0.5652173913043478,
    "test_f1_semicolon": 0.55,
    "test_f1_colon": 0.48120300751879697,
    "test_f1_hyphen": 0.17647058823529413,
    "test_runtime": 4.8056,
    "test_samples_per_second": 121.108,
    "test_steps_per_second": 2.081
  }
}