trainer_state.json

{
  "best_metric": 0.8594666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0817/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": 7e-05,
      "loss": 1.6399,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7824,
      "eval_loss": 0.7512590289115906,
      "eval_runtime": 11.466,
      "eval_samples_per_second": 327.054,
      "eval_steps_per_second": 5.146,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.537,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8277333333333333,
      "eval_loss": 0.5704519152641296,
      "eval_runtime": 10.8778,
      "eval_samples_per_second": 344.737,
      "eval_steps_per_second": 5.424,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.870721817016602,
      "learning_rate": 7e-05,
      "loss": 0.331,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8418666666666667,
      "eval_loss": 0.5140174627304077,
      "eval_runtime": 11.1203,
      "eval_samples_per_second": 337.221,
      "eval_steps_per_second": 5.306,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 14.227957725524902,
      "learning_rate": 7e-05,
      "loss": 0.233,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.49855855107307434,
      "eval_runtime": 11.5599,
      "eval_samples_per_second": 324.396,
      "eval_steps_per_second": 5.104,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.5817592144012451,
      "learning_rate": 7e-05,
      "loss": 0.1658,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8389333333333333,
      "eval_loss": 0.5570365190505981,
      "eval_runtime": 11.8179,
      "eval_samples_per_second": 317.316,
      "eval_steps_per_second": 4.992,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 15.709019660949707,
      "learning_rate": 7e-05,
      "loss": 0.127,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.5181446075439453,
      "eval_runtime": 11.2084,
      "eval_samples_per_second": 334.571,
      "eval_steps_per_second": 5.264,
      "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
}