trainer_state.json

{
  "best_metric": 0.9964285714285714,
  "best_model_checkpoint": "vit-large-patch16-224-dungeon-geo-morphs-0-4-30Nov24-015/checkpoint-50",
  "epoch": 23.90909090909091,
  "eval_steps": 10,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.909090909090909,
      "grad_norm": 22.980161666870117,
      "learning_rate": 1.888888888888889e-05,
      "loss": 1.4584,
      "step": 10
    },
    {
      "epoch": 3.909090909090909,
      "eval_accuracy": 0.675,
      "eval_loss": 1.0524959564208984,
      "eval_runtime": 7.6054,
      "eval_samples_per_second": 73.631,
      "eval_steps_per_second": 9.204,
      "step": 10
    },
    {
      "epoch": 7.909090909090909,
      "grad_norm": 21.107227325439453,
      "learning_rate": 1.5185185185185187e-05,
      "loss": 0.5841,
      "step": 20
    },
    {
      "epoch": 7.909090909090909,
      "eval_accuracy": 0.9375,
      "eval_loss": 0.4200492799282074,
      "eval_runtime": 8.5988,
      "eval_samples_per_second": 65.125,
      "eval_steps_per_second": 8.141,
      "step": 20
    },
    {
      "epoch": 11.909090909090908,
      "grad_norm": 22.844900131225586,
      "learning_rate": 1.1481481481481482e-05,
      "loss": 0.2335,
      "step": 30
    },
    {
      "epoch": 11.909090909090908,
      "eval_accuracy": 0.9339285714285714,
      "eval_loss": 0.27219241857528687,
      "eval_runtime": 7.3355,
      "eval_samples_per_second": 76.341,
      "eval_steps_per_second": 9.543,
      "step": 30
    },
    {
      "epoch": 15.909090909090908,
      "grad_norm": 13.868156433105469,
      "learning_rate": 7.77777777777778e-06,
      "loss": 0.1077,
      "step": 40
    },
    {
      "epoch": 15.909090909090908,
      "eval_accuracy": 0.9767857142857143,
      "eval_loss": 0.16271629929542542,
      "eval_runtime": 8.3021,
      "eval_samples_per_second": 67.453,
      "eval_steps_per_second": 8.432,
      "step": 40
    },
    {
      "epoch": 19.90909090909091,
      "grad_norm": 10.624994277954102,
      "learning_rate": 4.074074074074074e-06,
      "loss": 0.0607,
      "step": 50
    },
    {
      "epoch": 19.90909090909091,
      "eval_accuracy": 0.9964285714285714,
      "eval_loss": 0.10052422434091568,
      "eval_runtime": 8.2,
      "eval_samples_per_second": 68.292,
      "eval_steps_per_second": 8.537,
      "step": 50
    },
    {
      "epoch": 23.90909090909091,
      "grad_norm": 4.887649059295654,
      "learning_rate": 3.7037037037037036e-07,
      "loss": 0.0273,
      "step": 60
    },
    {
      "epoch": 23.90909090909091,
      "eval_accuracy": 0.9946428571428572,
      "eval_loss": 0.09675189107656479,
      "eval_runtime": 8.8105,
      "eval_samples_per_second": 63.561,
      "eval_steps_per_second": 7.945,
      "step": 60
    },
    {
      "epoch": 23.90909090909091,
      "step": 60,
      "total_flos": 5.259589576438579e+17,
      "train_loss": 0.41194454034169514,
      "train_runtime": 616.1526,
      "train_samples_per_second": 3.993,
      "train_steps_per_second": 0.097
    }
  ],
  "logging_steps": 10,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.259589576438579e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}