trainer_state.json

{
  "best_metric": 0.9442666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0944/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": 2.0539357662200928,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 1.1978,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9285333333333333,
      "eval_loss": 0.3150501847267151,
      "eval_runtime": 6.9314,
      "eval_samples_per_second": 541.017,
      "eval_steps_per_second": 8.512,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.0933971405029297,
      "learning_rate": 2.649066664678467e-05,
      "loss": 0.1653,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9357333333333333,
      "eval_loss": 0.24109531939029694,
      "eval_runtime": 6.8358,
      "eval_samples_per_second": 548.582,
      "eval_steps_per_second": 8.631,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.15625926852226257,
      "learning_rate": 2.25e-05,
      "loss": 0.0582,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9421333333333334,
      "eval_loss": 0.2266910821199417,
      "eval_runtime": 6.7385,
      "eval_samples_per_second": 556.504,
      "eval_steps_per_second": 8.756,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5203189849853516,
      "learning_rate": 1.760472266500396e-05,
      "loss": 0.0214,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9426666666666667,
      "eval_loss": 0.2231871485710144,
      "eval_runtime": 6.6684,
      "eval_samples_per_second": 562.351,
      "eval_steps_per_second": 8.848,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.058746419847011566,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 0.0079,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9432,
      "eval_loss": 0.2350272387266159,
      "eval_runtime": 6.7886,
      "eval_samples_per_second": 552.398,
      "eval_steps_per_second": 8.691,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.08836039900779724,
      "learning_rate": 7.500000000000004e-06,
      "loss": 0.0051,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9442666666666667,
      "eval_loss": 0.2292039543390274,
      "eval_runtime": 7.0534,
      "eval_samples_per_second": 531.662,
      "eval_steps_per_second": 8.365,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}