trainer_state.json

{
  "best_metric": 0.6089015603065491,
  "best_model_checkpoint": "./vit-base-brain-mri\\checkpoint-648",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 720,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6289198606271778,
      "eval_loss": 0.9818006753921509,
      "eval_runtime": 4.8676,
      "eval_samples_per_second": 117.922,
      "eval_steps_per_second": 14.792,
      "step": 72
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 2.2408666610717773,
      "learning_rate": 0.00021666666666666666,
      "loss": 1.0744,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.686411149825784,
      "eval_loss": 0.8286930322647095,
      "eval_runtime": 4.7946,
      "eval_samples_per_second": 119.718,
      "eval_steps_per_second": 15.017,
      "step": 144
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 4.1483378410339355,
      "learning_rate": 0.0001333333333333333,
      "loss": 0.7716,
      "step": 200
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7160278745644599,
      "eval_loss": 0.7534749507904053,
      "eval_runtime": 4.9187,
      "eval_samples_per_second": 116.696,
      "eval_steps_per_second": 14.638,
      "step": 216
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.740418118466899,
      "eval_loss": 0.7139923572540283,
      "eval_runtime": 4.8084,
      "eval_samples_per_second": 119.374,
      "eval_steps_per_second": 14.974,
      "step": 288
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 3.2614974975585938,
      "learning_rate": 4.9999999999999996e-05,
      "loss": 0.6975,
      "step": 300
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7491289198606271,
      "eval_loss": 0.7015281915664673,
      "eval_runtime": 4.8073,
      "eval_samples_per_second": 119.403,
      "eval_steps_per_second": 14.977,
      "step": 360
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 1.9345481395721436,
      "learning_rate": 0.0001333333333333333,
      "loss": 0.6651,
      "step": 400
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7630662020905923,
      "eval_loss": 0.68391352891922,
      "eval_runtime": 4.8866,
      "eval_samples_per_second": 117.465,
      "eval_steps_per_second": 14.734,
      "step": 432
    },
    {
      "epoch": 6.944444444444445,
      "grad_norm": 2.201732635498047,
      "learning_rate": 9.166666666666667e-05,
      "loss": 0.6307,
      "step": 500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7700348432055749,
      "eval_loss": 0.6623685359954834,
      "eval_runtime": 4.9432,
      "eval_samples_per_second": 116.118,
      "eval_steps_per_second": 14.565,
      "step": 504
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7822299651567944,
      "eval_loss": 0.6363218426704407,
      "eval_runtime": 4.9241,
      "eval_samples_per_second": 116.57,
      "eval_steps_per_second": 14.622,
      "step": 576
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 1.8554637432098389,
      "learning_rate": 4.9999999999999996e-05,
      "loss": 0.5857,
      "step": 600
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7822299651567944,
      "eval_loss": 0.6089015603065491,
      "eval_runtime": 4.9259,
      "eval_samples_per_second": 116.527,
      "eval_steps_per_second": 14.617,
      "step": 648
    },
    {
      "epoch": 9.722222222222221,
      "grad_norm": 3.4061734676361084,
      "learning_rate": 8.333333333333332e-06,
      "loss": 0.576,
      "step": 700
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7770034843205574,
      "eval_loss": 0.6249111294746399,
      "eval_runtime": 5.1184,
      "eval_samples_per_second": 112.144,
      "eval_steps_per_second": 14.067,
      "step": 720
    },
    {
      "epoch": 10.0,
      "step": 720,
      "total_flos": 1.779247974652969e+18,
      "train_loss": 0.3026825189590454,
      "train_runtime": 130.7944,
      "train_samples_per_second": 175.543,
      "train_steps_per_second": 5.505
    }
  ],
  "logging_steps": 100,
  "max_steps": 720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.779247974652969e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}