trainer_state.json

{
  "best_metric": 0.6021333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0456/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 0.0004850126564054637,
      "loss": 3.0799,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.26266666666666666,
      "eval_loss": 2.6985745429992676,
      "eval_runtime": 5.3353,
      "eval_samples_per_second": 702.868,
      "eval_steps_per_second": 11.058,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00044184758936602033,
      "loss": 2.46,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.37093333333333334,
      "eval_loss": 2.207042694091797,
      "eval_runtime": 5.1949,
      "eval_samples_per_second": 721.862,
      "eval_steps_per_second": 11.357,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.9866108894348145,
      "learning_rate": 0.0003754536287075859,
      "loss": 2.0081,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.44853333333333334,
      "eval_loss": 1.9422286748886108,
      "eval_runtime": 5.4058,
      "eval_samples_per_second": 693.702,
      "eval_steps_per_second": 10.914,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 15.059794425964355,
      "learning_rate": 0.000293928108928647,
      "loss": 1.6546,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5088,
      "eval_loss": 1.722028136253357,
      "eval_runtime": 5.3814,
      "eval_samples_per_second": 696.842,
      "eval_steps_per_second": 10.964,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.605114698410034,
      "learning_rate": 0.00020710421090307243,
      "loss": 1.2661,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5530666666666667,
      "eval_loss": 1.5556328296661377,
      "eval_runtime": 4.8254,
      "eval_samples_per_second": 777.146,
      "eval_steps_per_second": 12.227,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 12.404691696166992,
      "learning_rate": 0.00012545417811701015,
      "loss": 0.8652,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5778666666666666,
      "eval_loss": 1.4948747158050537,
      "eval_runtime": 4.7719,
      "eval_samples_per_second": 785.855,
      "eval_steps_per_second": 12.364,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.777204990386963,
      "learning_rate": 5.882620955060414e-05,
      "loss": 0.4799,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5909333333333333,
      "eval_loss": 1.4885900020599365,
      "eval_runtime": 4.3646,
      "eval_samples_per_second": 859.19,
      "eval_steps_per_second": 13.518,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 5.8714213371276855,
      "learning_rate": 1.5256621537052151e-05,
      "loss": 0.2343,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6021333333333333,
      "eval_loss": 1.502152442932129,
      "eval_runtime": 4.4067,
      "eval_samples_per_second": 850.976,
      "eval_steps_per_second": 13.389,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}