trainer_state.json

{
  "best_metric": 0.717948717948718,
  "best_model_checkpoint": "swinv2-tiny-patch4-window16-256-prueba3/checkpoint-11",
  "epoch": 2.7272727272727275,
  "eval_steps": 500,
  "global_step": 15,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9090909090909091,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.616500735282898,
      "eval_runtime": 0.7044,
      "eval_samples_per_second": 110.728,
      "eval_steps_per_second": 4.259,
      "step": 5
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.8319458961486816,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.6601,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.717948717948718,
      "eval_loss": 0.5599234700202942,
      "eval_runtime": 0.7351,
      "eval_samples_per_second": 106.109,
      "eval_steps_per_second": 4.081,
      "step": 11
    },
    {
      "epoch": 2.7272727272727275,
      "eval_accuracy": 0.717948717948718,
      "eval_loss": 0.5461894869804382,
      "eval_runtime": 0.6997,
      "eval_samples_per_second": 111.469,
      "eval_steps_per_second": 4.287,
      "step": 15
    },
    {
      "epoch": 2.7272727272727275,
      "step": 15,
      "total_flos": 4.767600833578598e+16,
      "train_loss": 0.6222841421763102,
      "train_runtime": 42.1958,
      "train_samples_per_second": 49.839,
      "train_steps_per_second": 0.355
    }
  ],
  "logging_steps": 10,
  "max_steps": 15,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 4.767600833578598e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}