trainer_state.json

{
  "best_metric": 0.8309333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0254/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 1.4611,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7725333333333333,
      "eval_loss": 0.7448643445968628,
      "eval_runtime": 5.2961,
      "eval_samples_per_second": 708.067,
      "eval_steps_per_second": 11.14,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.095996379852295,
      "learning_rate": 0.0003,
      "loss": 0.6191,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7832,
      "eval_loss": 0.6817769408226013,
      "eval_runtime": 4.2804,
      "eval_samples_per_second": 876.094,
      "eval_steps_per_second": 13.784,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.15286697447299957,
      "learning_rate": 0.0003,
      "loss": 0.4305,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6263529062271118,
      "eval_runtime": 5.53,
      "eval_samples_per_second": 678.12,
      "eval_steps_per_second": 10.669,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.3203,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8029333333333334,
      "eval_loss": 0.6742420792579651,
      "eval_runtime": 6.256,
      "eval_samples_per_second": 599.428,
      "eval_steps_per_second": 9.431,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.8071006536483765,
      "learning_rate": 0.0003,
      "loss": 0.261,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8058666666666666,
      "eval_loss": 0.6851070523262024,
      "eval_runtime": 7.7864,
      "eval_samples_per_second": 481.609,
      "eval_steps_per_second": 7.577,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 19.037029266357422,
      "learning_rate": 0.0003,
      "loss": 0.213,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8309333333333333,
      "eval_loss": 0.6060481667518616,
      "eval_runtime": 4.6069,
      "eval_samples_per_second": 813.989,
      "eval_steps_per_second": 12.807,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}