trainer_state.json

{
  "best_metric": 0.004365460481494665,
  "best_model_checkpoint": "./vit-brain-tumour-v1/checkpoint-100",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 148,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 2.4590463638305664,
      "learning_rate": 0.0001864864864864865,
      "loss": 0.4827,
      "step": 10
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 2.9085402488708496,
      "learning_rate": 0.000172972972972973,
      "loss": 0.221,
      "step": 20
    },
    {
      "epoch": 0.8108108108108109,
      "grad_norm": 0.13484928011894226,
      "learning_rate": 0.00015945945945945947,
      "loss": 0.0573,
      "step": 30
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 3.143681287765503,
      "learning_rate": 0.00014594594594594595,
      "loss": 0.0655,
      "step": 40
    },
    {
      "epoch": 1.3513513513513513,
      "grad_norm": 1.2873934507369995,
      "learning_rate": 0.00013243243243243243,
      "loss": 0.066,
      "step": 50
    },
    {
      "epoch": 1.6216216216216215,
      "grad_norm": 0.05828818306326866,
      "learning_rate": 0.00011891891891891893,
      "loss": 0.009,
      "step": 60
    },
    {
      "epoch": 1.8918918918918919,
      "grad_norm": 0.04554607346653938,
      "learning_rate": 0.0001054054054054054,
      "loss": 0.0063,
      "step": 70
    },
    {
      "epoch": 2.1621621621621623,
      "grad_norm": 0.0411420576274395,
      "learning_rate": 9.18918918918919e-05,
      "loss": 0.0053,
      "step": 80
    },
    {
      "epoch": 2.4324324324324325,
      "grad_norm": 0.03753970190882683,
      "learning_rate": 7.837837837837838e-05,
      "loss": 0.0048,
      "step": 90
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 0.03439953923225403,
      "learning_rate": 6.486486486486487e-05,
      "loss": 0.0043,
      "step": 100
    },
    {
      "epoch": 2.7027027027027026,
      "eval_accuracy": 1.0,
      "eval_loss": 0.004365460481494665,
      "eval_runtime": 2.8872,
      "eval_samples_per_second": 50.222,
      "eval_steps_per_second": 6.581,
      "step": 100
    },
    {
      "epoch": 2.972972972972973,
      "grad_norm": 0.032108623534440994,
      "learning_rate": 5.135135135135135e-05,
      "loss": 0.0041,
      "step": 110
    },
    {
      "epoch": 3.2432432432432434,
      "grad_norm": 0.03147277981042862,
      "learning_rate": 3.783783783783784e-05,
      "loss": 0.0038,
      "step": 120
    },
    {
      "epoch": 3.5135135135135136,
      "grad_norm": 0.02981223724782467,
      "learning_rate": 2.4324324324324327e-05,
      "loss": 0.0037,
      "step": 130
    },
    {
      "epoch": 3.7837837837837838,
      "grad_norm": 0.03125371038913727,
      "learning_rate": 1.0810810810810812e-05,
      "loss": 0.0037,
      "step": 140
    },
    {
      "epoch": 4.0,
      "step": 148,
      "total_flos": 1.800913838643118e+17,
      "train_loss": 0.06353814376367105,
      "train_runtime": 74.4207,
      "train_samples_per_second": 31.228,
      "train_steps_per_second": 1.989
    }
  ],
  "logging_steps": 10,
  "max_steps": 148,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 1.800913838643118e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}