trainer_state.json

{
  "best_metric": 0.5275640487670898,
  "best_model_checkpoint": "vit-base-patch16-224-in21k-rotated-dungeons-v4/checkpoint-100",
  "epoch": 15.0,
  "eval_steps": 100,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.25,
      "grad_norm": 0.5645539164543152,
      "learning_rate": 0.00018333333333333334,
      "loss": 0.1955,
      "step": 10
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.22394521534442902,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.0944,
      "step": 20
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.37107622623443604,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.0686,
      "step": 30
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.1673012524843216,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.0459,
      "step": 40
    },
    {
      "epoch": 6.25,
      "grad_norm": 0.12259427458047867,
      "learning_rate": 0.00011666666666666668,
      "loss": 0.042,
      "step": 50
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.1604679673910141,
      "learning_rate": 0.0001,
      "loss": 0.0327,
      "step": 60
    },
    {
      "epoch": 8.75,
      "grad_norm": 0.121036596596241,
      "learning_rate": 8.333333333333334e-05,
      "loss": 0.0336,
      "step": 70
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.12557414174079895,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.0295,
      "step": 80
    },
    {
      "epoch": 11.25,
      "grad_norm": 0.10457774996757507,
      "learning_rate": 5e-05,
      "loss": 0.0266,
      "step": 90
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.10654424130916595,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.028,
      "step": 100
    },
    {
      "epoch": 12.5,
      "eval_accuracy": 0.875,
      "eval_loss": 0.5275640487670898,
      "eval_runtime": 0.0716,
      "eval_samples_per_second": 111.698,
      "eval_steps_per_second": 13.962,
      "step": 100
    },
    {
      "epoch": 13.75,
      "grad_norm": 0.17629727721214294,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.028,
      "step": 110
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.14613842964172363,
      "learning_rate": 0.0,
      "loss": 0.0244,
      "step": 120
    },
    {
      "epoch": 15.0,
      "step": 120,
      "total_flos": 4.64972773773312e+16,
      "train_loss": 0.05409030864636103,
      "train_runtime": 24.6133,
      "train_samples_per_second": 24.377,
      "train_steps_per_second": 4.875
    }
  ],
  "logging_steps": 10,
  "max_steps": 120,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.64972773773312e+16,
  "train_batch_size": 5,
  "trial_name": null,
  "trial_params": null
}