trainer_state.json

{
  "best_metric": 0.8368,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0711/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": 10.183516502380371,
      "learning_rate": 0.0005,
      "loss": 0.8322,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8096,
      "eval_loss": 0.6360242366790771,
      "eval_runtime": 13.0778,
      "eval_samples_per_second": 286.746,
      "eval_steps_per_second": 4.511,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.37233853340148926,
      "learning_rate": 0.0005,
      "loss": 0.5166,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8016,
      "eval_loss": 0.7077659964561462,
      "eval_runtime": 13.4527,
      "eval_samples_per_second": 278.755,
      "eval_steps_per_second": 4.386,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.885197639465332,
      "learning_rate": 0.0005,
      "loss": 0.4032,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7973333333333333,
      "eval_loss": 0.7441632747650146,
      "eval_runtime": 13.4944,
      "eval_samples_per_second": 277.893,
      "eval_steps_per_second": 4.372,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.13954609632492065,
      "learning_rate": 0.0005,
      "loss": 0.3417,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8176,
      "eval_loss": 0.6623221635818481,
      "eval_runtime": 13.3921,
      "eval_samples_per_second": 280.016,
      "eval_steps_per_second": 4.406,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.028029752895236015,
      "learning_rate": 0.0005,
      "loss": 0.292,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8304,
      "eval_loss": 0.6049489378929138,
      "eval_runtime": 13.183,
      "eval_samples_per_second": 284.457,
      "eval_steps_per_second": 4.475,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.013776368461549282,
      "learning_rate": 0.0005,
      "loss": 0.2539,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8264,
      "eval_loss": 0.6359582543373108,
      "eval_runtime": 13.3452,
      "eval_samples_per_second": 281.0,
      "eval_steps_per_second": 4.421,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 0.0005,
      "loss": 0.2309,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8024,
      "eval_loss": 0.7783176302909851,
      "eval_runtime": 13.5159,
      "eval_samples_per_second": 277.451,
      "eval_steps_per_second": 4.365,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.37624362111091614,
      "learning_rate": 0.0005,
      "loss": 0.214,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8368,
      "eval_loss": 0.6329951286315918,
      "eval_runtime": 12.8393,
      "eval_samples_per_second": 292.071,
      "eval_steps_per_second": 4.595,
      "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
}