trainer_state.json

{
  "best_metric": 0.9801587301587301,
  "best_model_checkpoint": "vit-base-patch16-224-in21k-eurosat/checkpoint-132",
  "epoch": 4.977777777777778,
  "global_step": 165,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3,
      "learning_rate": 4.1216216216216216e-05,
      "loss": 1.1013,
      "step": 10
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.783783783783784e-05,
      "loss": 0.8777,
      "step": 20
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.445945945945946e-05,
      "loss": 0.7076,
      "step": 30
    },
    {
      "epoch": 0.98,
      "eval_accuracy": 0.9695767195767195,
      "eval_loss": 0.6118746995925903,
      "eval_runtime": 9.7719,
      "eval_samples_per_second": 77.365,
      "eval_steps_per_second": 2.456,
      "step": 33
    },
    {
      "epoch": 1.21,
      "learning_rate": 3.108108108108108e-05,
      "loss": 0.6394,
      "step": 40
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.7702702702702704e-05,
      "loss": 0.5029,
      "step": 50
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.4324324324324327e-05,
      "loss": 0.4469,
      "step": 60
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.9788359788359788,
      "eval_loss": 0.4189939498901367,
      "eval_runtime": 9.9093,
      "eval_samples_per_second": 76.292,
      "eval_steps_per_second": 2.422,
      "step": 66
    },
    {
      "epoch": 2.12,
      "learning_rate": 2.0945945945945947e-05,
      "loss": 0.4382,
      "step": 70
    },
    {
      "epoch": 2.41,
      "learning_rate": 1.756756756756757e-05,
      "loss": 0.3635,
      "step": 80
    },
    {
      "epoch": 2.71,
      "learning_rate": 1.4189189189189189e-05,
      "loss": 0.3497,
      "step": 90
    },
    {
      "epoch": 2.98,
      "eval_accuracy": 0.9788359788359788,
      "eval_loss": 0.355459064245224,
      "eval_runtime": 9.8101,
      "eval_samples_per_second": 77.064,
      "eval_steps_per_second": 2.446,
      "step": 99
    },
    {
      "epoch": 3.03,
      "learning_rate": 1.0810810810810812e-05,
      "loss": 0.3641,
      "step": 100
    },
    {
      "epoch": 3.33,
      "learning_rate": 7.432432432432433e-06,
      "loss": 0.3099,
      "step": 110
    },
    {
      "epoch": 3.62,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 0.3016,
      "step": 120
    },
    {
      "epoch": 3.92,
      "learning_rate": 6.756756756756758e-07,
      "loss": 0.3048,
      "step": 130
    },
    {
      "epoch": 3.98,
      "eval_accuracy": 0.9801587301587301,
      "eval_loss": 0.33940190076828003,
      "eval_runtime": 9.7751,
      "eval_samples_per_second": 77.339,
      "eval_steps_per_second": 2.455,
      "step": 132
    },
    {
      "epoch": 4.24,
      "learning_rate": 0.0,
      "loss": 0.3238,
      "step": 140
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.0,
      "loss": 0.2969,
      "step": 150
    },
    {
      "epoch": 4.83,
      "learning_rate": 0.0,
      "loss": 0.2983,
      "step": 160
    },
    {
      "epoch": 4.98,
      "eval_accuracy": 0.9801587301587301,
      "eval_loss": 0.33940190076828003,
      "eval_runtime": 9.7782,
      "eval_samples_per_second": 77.315,
      "eval_steps_per_second": 2.454,
      "step": 165
    },
    {
      "epoch": 4.98,
      "step": 165,
      "total_flos": 1.6665072271223685e+18,
      "train_loss": 0.47145734555793534,
      "train_runtime": 824.3744,
      "train_samples_per_second": 26.202,
      "train_steps_per_second": 0.2
    }
  ],
  "max_steps": 165,
  "num_train_epochs": 5,
  "total_flos": 1.6665072271223685e+18,
  "trial_name": null,
  "trial_params": null
}