trainer_state.json

{
  "best_metric": 0.8128,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0507/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": 1.5515,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7418666666666667,
      "eval_loss": 0.8300451636314392,
      "eval_runtime": 6.5588,
      "eval_samples_per_second": 571.75,
      "eval_steps_per_second": 8.996,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.046992301940918,
      "learning_rate": 0.0003,
      "loss": 0.6474,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7565333333333333,
      "eval_loss": 0.7560407519340515,
      "eval_runtime": 7.045,
      "eval_samples_per_second": 532.293,
      "eval_steps_per_second": 8.375,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 10.275901794433594,
      "learning_rate": 0.0003,
      "loss": 0.4531,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7978666666666666,
      "eval_loss": 0.6515294909477234,
      "eval_runtime": 6.6196,
      "eval_samples_per_second": 566.496,
      "eval_steps_per_second": 8.913,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5656812191009521,
      "learning_rate": 0.0003,
      "loss": 0.3542,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7837333333333333,
      "eval_loss": 0.7066212296485901,
      "eval_runtime": 6.4246,
      "eval_samples_per_second": 583.696,
      "eval_steps_per_second": 9.183,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.05112788826227188,
      "learning_rate": 0.0003,
      "loss": 0.2771,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.808,
      "eval_loss": 0.6646728515625,
      "eval_runtime": 6.5226,
      "eval_samples_per_second": 574.922,
      "eval_steps_per_second": 9.045,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.5691249370574951,
      "learning_rate": 0.0003,
      "loss": 0.2227,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8128,
      "eval_loss": 0.6249762177467346,
      "eval_runtime": 6.6436,
      "eval_samples_per_second": 564.453,
      "eval_steps_per_second": 8.881,
      "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
}