checkpoint-54/trainer_state.json

{
  "best_metric": 0.8523994617102314,
  "best_model_checkpoint": "distilhubert-finetuned-mixed-data/checkpoint-54",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 54,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8888888888888888,
      "eval_accuracy": 0.63003663003663,
      "eval_confusion_matrix": [
        [
          53,
          0,
          16,
          3
        ],
        [
          28,
          1,
          31,
          0
        ],
        [
          13,
          0,
          62,
          1
        ],
        [
          3,
          0,
          6,
          56
        ]
      ],
      "eval_f1": 0.566693372031096,
      "eval_loss": 1.31327486038208,
      "eval_precision": 0.7361933549293478,
      "eval_recall": 0.63003663003663,
      "eval_runtime": 2.8436,
      "eval_samples_per_second": 96.005,
      "eval_steps_per_second": 1.055,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6593406593406593,
      "eval_confusion_matrix": [
        [
          41,
          10,
          16,
          5
        ],
        [
          21,
          1,
          38,
          0
        ],
        [
          1,
          2,
          73,
          0
        ],
        [
          0,
          0,
          0,
          65
        ]
      ],
      "eval_f1": 0.595713773130153,
      "eval_loss": 0.9242589473724365,
      "eval_precision": 0.5696509512811321,
      "eval_recall": 0.6593406593406593,
      "eval_runtime": 2.8668,
      "eval_samples_per_second": 95.227,
      "eval_steps_per_second": 1.046,
      "step": 9
    },
    {
      "epoch": 2.888888888888889,
      "eval_accuracy": 0.717948717948718,
      "eval_confusion_matrix": [
        [
          52,
          8,
          7,
          5
        ],
        [
          18,
          5,
          37,
          0
        ],
        [
          1,
          0,
          75,
          0
        ],
        [
          1,
          0,
          0,
          64
        ]
      ],
      "eval_f1": 0.6621621567125869,
      "eval_loss": 0.8232662081718445,
      "eval_precision": 0.671303801513745,
      "eval_recall": 0.717948717948718,
      "eval_runtime": 2.8711,
      "eval_samples_per_second": 95.084,
      "eval_steps_per_second": 1.045,
      "step": 13
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6959706959706959,
      "eval_confusion_matrix": [
        [
          45,
          9,
          13,
          5
        ],
        [
          13,
          5,
          42,
          0
        ],
        [
          1,
          0,
          75,
          0
        ],
        [
          0,
          0,
          0,
          65
        ]
      ],
      "eval_f1": 0.6428796900399459,
      "eval_loss": 0.951453447341919,
      "eval_precision": 0.6613447705829899,
      "eval_recall": 0.6959706959706959,
      "eval_runtime": 2.8813,
      "eval_samples_per_second": 94.75,
      "eval_steps_per_second": 1.041,
      "step": 18
    },
    {
      "epoch": 4.888888888888889,
      "eval_accuracy": 0.6666666666666666,
      "eval_confusion_matrix": [
        [
          16,
          49,
          3,
          4
        ],
        [
          1,
          26,
          33,
          0
        ],
        [
          0,
          1,
          75,
          0
        ],
        [
          0,
          0,
          0,
          65
        ]
      ],
      "eval_f1": 0.6331541584523206,
      "eval_loss": 1.0080645084381104,
      "eval_precision": 0.7358031394913882,
      "eval_recall": 0.6666666666666666,
      "eval_runtime": 2.9108,
      "eval_samples_per_second": 93.789,
      "eval_steps_per_second": 1.031,
      "step": 22
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.652014652014652,
      "eval_confusion_matrix": [
        [
          66,
          0,
          0,
          6
        ],
        [
          36,
          16,
          8,
          0
        ],
        [
          10,
          35,
          31,
          0
        ],
        [
          0,
          0,
          0,
          65
        ]
      ],
      "eval_f1": 0.6302407955860642,
      "eval_loss": 0.9598046541213989,
      "eval_precision": 0.6636241740077424,
      "eval_recall": 0.652014652014652,
      "eval_runtime": 2.9773,
      "eval_samples_per_second": 91.693,
      "eval_steps_per_second": 1.008,
      "step": 27
    },
    {
      "epoch": 6.888888888888889,
      "eval_accuracy": 0.7692307692307693,
      "eval_confusion_matrix": [
        [
          66,
          0,
          0,
          6
        ],
        [
          29,
          21,
          10,
          0
        ],
        [
          3,
          15,
          58,
          0
        ],
        [
          0,
          0,
          0,
          65
        ]
      ],
      "eval_f1": 0.7527855586679116,
      "eval_loss": 0.7898163795471191,
      "eval_precision": 0.7612467677056892,
      "eval_recall": 0.7692307692307693,
      "eval_runtime": 2.8918,
      "eval_samples_per_second": 94.404,
      "eval_steps_per_second": 1.037,
      "step": 31
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7802197802197802,
      "eval_confusion_matrix": [
        [
          66,
          0,
          0,
          6
        ],
        [
          25,
          15,
          19,
          1
        ],
        [
          1,
          6,
          68,
          1
        ],
        [
          1,
          0,
          0,
          64
        ]
      ],
      "eval_f1": 0.7471179200524057,
      "eval_loss": 0.7337484359741211,
      "eval_precision": 0.7733845922309294,
      "eval_recall": 0.7802197802197802,
      "eval_runtime": 2.9006,
      "eval_samples_per_second": 94.119,
      "eval_steps_per_second": 1.034,
      "step": 36
    },
    {
      "epoch": 8.88888888888889,
      "eval_accuracy": 0.7912087912087912,
      "eval_confusion_matrix": [
        [
          57,
          8,
          3,
          4
        ],
        [
          13,
          24,
          23,
          0
        ],
        [
          0,
          5,
          71,
          0
        ],
        [
          1,
          0,
          0,
          64
        ]
      ],
      "eval_f1": 0.7766564722186922,
      "eval_loss": 0.7148727178573608,
      "eval_precision": 0.7821504483074875,
      "eval_recall": 0.7912087912087912,
      "eval_runtime": 2.8866,
      "eval_samples_per_second": 94.576,
      "eval_steps_per_second": 1.039,
      "step": 40
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8168498168498168,
      "eval_confusion_matrix": [
        [
          51,
          8,
          1,
          12
        ],
        [
          10,
          36,
          14,
          0
        ],
        [
          0,
          5,
          71,
          0
        ],
        [
          0,
          0,
          0,
          65
        ]
      ],
      "eval_f1": 0.8094336953840884,
      "eval_loss": 0.7574812173843384,
      "eval_precision": 0.8127936625684181,
      "eval_recall": 0.8168498168498168,
      "eval_runtime": 2.8967,
      "eval_samples_per_second": 94.245,
      "eval_steps_per_second": 1.036,
      "step": 45
    },
    {
      "epoch": 10.88888888888889,
      "eval_accuracy": 0.8021978021978022,
      "eval_confusion_matrix": [
        [
          48,
          18,
          2,
          4
        ],
        [
          4,
          41,
          15,
          0
        ],
        [
          0,
          7,
          69,
          0
        ],
        [
          3,
          0,
          1,
          61
        ]
      ],
      "eval_f1": 0.801525180147331,
      "eval_loss": 0.7140281200408936,
      "eval_precision": 0.8109338936925145,
      "eval_recall": 0.8021978021978022,
      "eval_runtime": 2.917,
      "eval_samples_per_second": 93.589,
      "eval_steps_per_second": 1.028,
      "step": 49
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8534798534798534,
      "eval_confusion_matrix": [
        [
          56,
          12,
          0,
          4
        ],
        [
          6,
          43,
          11,
          0
        ],
        [
          0,
          6,
          70,
          0
        ],
        [
          1,
          0,
          0,
          64
        ]
      ],
      "eval_f1": 0.8523994617102314,
      "eval_loss": 0.6672152280807495,
      "eval_precision": 0.8540313732642031,
      "eval_recall": 0.8534798534798534,
      "eval_runtime": 2.8969,
      "eval_samples_per_second": 94.238,
      "eval_steps_per_second": 1.036,
      "step": 54
    }
  ],
  "logging_steps": 500,
  "max_steps": 180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 45,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.97468834048e+16,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}