trainer_state.json

{
  "best_metric": 0.8728,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0315/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": 9.621650897886543e-05,
      "loss": 1.6056,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7581333333333333,
      "eval_loss": 0.815542995929718,
      "eval_runtime": 4.38,
      "eval_samples_per_second": 856.169,
      "eval_steps_per_second": 13.47,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.067920207977295,
      "learning_rate": 8.539700818671957e-05,
      "loss": 0.5942,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.796,
      "eval_loss": 0.6835128664970398,
      "eval_runtime": 4.3323,
      "eval_samples_per_second": 865.589,
      "eval_steps_per_second": 13.619,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.42041704058647156,
      "learning_rate": 6.918863377282392e-05,
      "loss": 0.343,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8288,
      "eval_loss": 0.5502464771270752,
      "eval_runtime": 4.3629,
      "eval_samples_per_second": 859.529,
      "eval_steps_per_second": 13.523,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.22988183796405792,
      "learning_rate": 5.005896381241413e-05,
      "loss": 0.1879,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.514340877532959,
      "eval_runtime": 5.2439,
      "eval_samples_per_second": 715.115,
      "eval_steps_per_second": 11.251,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.14875738322734833,
      "learning_rate": 3.0920317146072576e-05,
      "loss": 0.0924,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.5032705068588257,
      "eval_runtime": 5.4011,
      "eval_samples_per_second": 694.299,
      "eval_steps_per_second": 10.924,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.4767557978630066,
      "learning_rate": 1.4686379236485708e-05,
      "loss": 0.0403,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.4783107340335846,
      "eval_runtime": 4.5297,
      "eval_samples_per_second": 827.871,
      "eval_steps_per_second": 13.025,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.1589982509613037,
      "learning_rate": 3.82861996937452e-06,
      "loss": 0.0179,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8714666666666666,
      "eval_loss": 0.47757938504219055,
      "eval_runtime": 4.3975,
      "eval_samples_per_second": 852.764,
      "eval_steps_per_second": 13.417,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.5690861344337463,
      "learning_rate": 3.476732383100867e-11,
      "loss": 0.0114,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8728,
      "eval_loss": 0.46882036328315735,
      "eval_runtime": 5.7989,
      "eval_samples_per_second": 646.672,
      "eval_steps_per_second": 10.174,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}