trainer_state.json

{
  "best_global_step": 72,
  "best_metric": 0.55078125,
  "best_model_checkpoint": "convnext-tiny-224-finetuned-alzheimer-MRI/checkpoint-72",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 108,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 2.0088706016540527,
      "learning_rate": 4.0909090909090915e-05,
      "loss": 1.3197,
      "step": 10
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 2.71425199508667,
      "learning_rate": 4.5876288659793814e-05,
      "loss": 1.1091,
      "step": 20
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 1.3623636960983276,
      "learning_rate": 4.0721649484536084e-05,
      "loss": 1.0151,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5390625,
      "eval_loss": 0.9395096302032471,
      "eval_runtime": 3.0069,
      "eval_samples_per_second": 170.276,
      "eval_steps_per_second": 5.321,
      "step": 36
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 1.7469203472137451,
      "learning_rate": 3.5567010309278354e-05,
      "loss": 0.9697,
      "step": 40
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 1.5095868110656738,
      "learning_rate": 3.0412371134020617e-05,
      "loss": 0.9514,
      "step": 50
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 4.194490432739258,
      "learning_rate": 2.5257731958762887e-05,
      "loss": 0.9032,
      "step": 60
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 2.88313889503479,
      "learning_rate": 2.0103092783505157e-05,
      "loss": 0.9132,
      "step": 70
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.55078125,
      "eval_loss": 0.8595054745674133,
      "eval_runtime": 2.9709,
      "eval_samples_per_second": 172.338,
      "eval_steps_per_second": 5.386,
      "step": 72
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 2.767746686935425,
      "learning_rate": 1.4948453608247423e-05,
      "loss": 0.9126,
      "step": 80
    },
    {
      "epoch": 2.5,
      "grad_norm": 3.7601568698883057,
      "learning_rate": 9.793814432989691e-06,
      "loss": 0.884,
      "step": 90
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 3.920647382736206,
      "learning_rate": 4.639175257731959e-06,
      "loss": 0.876,
      "step": 100
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.548828125,
      "eval_loss": 0.8436535596847534,
      "eval_runtime": 2.9884,
      "eval_samples_per_second": 171.329,
      "eval_steps_per_second": 5.354,
      "step": 108
    },
    {
      "epoch": 3.0,
      "step": 108,
      "total_flos": 3.473836763020001e+17,
      "train_loss": 0.9776477946175469,
      "train_runtime": 256.1822,
      "train_samples_per_second": 53.962,
      "train_steps_per_second": 0.422
    }
  ],
  "logging_steps": 10,
  "max_steps": 108,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 3.473836763020001e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}