trainer_state.json

{
  "best_metric": 0.8989333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0988/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 2.85232922088503e-05,
      "loss": 2.1845,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8037333333333333,
      "eval_loss": 0.9650368094444275,
      "eval_runtime": 6.0191,
      "eval_samples_per_second": 623.021,
      "eval_steps_per_second": 9.802,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.732756614685059,
      "learning_rate": 2.4368144166557088e-05,
      "loss": 0.6403,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8616,
      "eval_loss": 0.5521084070205688,
      "eval_runtime": 6.2521,
      "eval_samples_per_second": 599.799,
      "eval_steps_per_second": 9.437,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.415581703186035,
      "learning_rate": 1.8357520279328874e-05,
      "loss": 0.3047,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8842666666666666,
      "eval_loss": 0.4306168556213379,
      "eval_runtime": 5.7981,
      "eval_samples_per_second": 646.761,
      "eval_steps_per_second": 10.176,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5187273025512695,
      "learning_rate": 1.1681898323495617e-05,
      "loss": 0.1529,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.3921929895877838,
      "eval_runtime": 5.7499,
      "eval_samples_per_second": 652.18,
      "eval_steps_per_second": 10.261,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3038378655910492,
      "learning_rate": 5.6634670985405755e-06,
      "loss": 0.0737,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.3650891184806824,
      "eval_runtime": 5.6759,
      "eval_samples_per_second": 660.683,
      "eval_steps_per_second": 10.395,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}