trainer_state.json

{
  "best_metric": 0.988479262672811,
  "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-gecko/checkpoint-210",
  "epoch": 9.655172413793103,
  "eval_steps": 500,
  "global_step": 210,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.97,
      "eval_accuracy": 0.6209677419354839,
      "eval_loss": 3.2698519229888916,
      "eval_runtime": 100.5477,
      "eval_samples_per_second": 8.633,
      "eval_steps_per_second": 0.278,
      "step": 21
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.8467741935483871,
      "eval_loss": 2.001060962677002,
      "eval_runtime": 86.2185,
      "eval_samples_per_second": 10.067,
      "eval_steps_per_second": 0.325,
      "step": 43
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.00042328042328042324,
      "loss": 3.1155,
      "step": 50
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.8640552995391705,
      "eval_loss": 1.2851452827453613,
      "eval_runtime": 96.6786,
      "eval_samples_per_second": 8.978,
      "eval_steps_per_second": 0.29,
      "step": 65
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9389400921658986,
      "eval_loss": 0.7751017808914185,
      "eval_runtime": 93.9452,
      "eval_samples_per_second": 9.239,
      "eval_steps_per_second": 0.298,
      "step": 87
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.000291005291005291,
      "loss": 1.1003,
      "step": 100
    },
    {
      "epoch": 4.97,
      "eval_accuracy": 0.9274193548387096,
      "eval_loss": 0.6059951186180115,
      "eval_runtime": 99.541,
      "eval_samples_per_second": 8.72,
      "eval_steps_per_second": 0.281,
      "step": 108
    },
    {
      "epoch": 5.98,
      "eval_accuracy": 0.9377880184331797,
      "eval_loss": 0.4584074318408966,
      "eval_runtime": 96.9396,
      "eval_samples_per_second": 8.954,
      "eval_steps_per_second": 0.289,
      "step": 130
    },
    {
      "epoch": 6.9,
      "learning_rate": 0.00015873015873015873,
      "loss": 0.5229,
      "step": 150
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.9585253456221198,
      "eval_loss": 0.3416719138622284,
      "eval_runtime": 97.2391,
      "eval_samples_per_second": 8.926,
      "eval_steps_per_second": 0.288,
      "step": 152
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9815668202764977,
      "eval_loss": 0.24149224162101746,
      "eval_runtime": 98.5205,
      "eval_samples_per_second": 8.81,
      "eval_steps_per_second": 0.284,
      "step": 174
    },
    {
      "epoch": 8.97,
      "eval_accuracy": 0.9873271889400922,
      "eval_loss": 0.2014480084180832,
      "eval_runtime": 95.774,
      "eval_samples_per_second": 9.063,
      "eval_steps_per_second": 0.292,
      "step": 195
    },
    {
      "epoch": 9.2,
      "learning_rate": 2.6455026455026453e-05,
      "loss": 0.3249,
      "step": 200
    },
    {
      "epoch": 9.66,
      "eval_accuracy": 0.988479262672811,
      "eval_loss": 0.18896923959255219,
      "eval_runtime": 97.6896,
      "eval_samples_per_second": 8.885,
      "eval_steps_per_second": 0.287,
      "step": 210
    },
    {
      "epoch": 9.66,
      "step": 210,
      "total_flos": 2.0674871916560548e+18,
      "train_loss": 1.218859517006647,
      "train_runtime": 3288.9398,
      "train_samples_per_second": 8.392,
      "train_steps_per_second": 0.064
    }
  ],
  "logging_steps": 50,
  "max_steps": 210,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.0674871916560548e+18,
  "trial_name": null,
  "trial_params": null
}