trainer_state.json

{
  "best_metric": 0.8853333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0646/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": NaN,
      "learning_rate": 7e-05,
      "loss": 0.7586,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8536,
      "eval_loss": 0.4813612997531891,
      "eval_runtime": 13.0035,
      "eval_samples_per_second": 288.383,
      "eval_steps_per_second": 4.537,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.33811983466148376,
      "learning_rate": 7e-05,
      "loss": 0.2443,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.4218961000442505,
      "eval_runtime": 12.4493,
      "eval_samples_per_second": 301.222,
      "eval_steps_per_second": 4.739,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.027779968455433846,
      "learning_rate": 7e-05,
      "loss": 0.1521,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.5624655485153198,
      "eval_runtime": 12.3769,
      "eval_samples_per_second": 302.984,
      "eval_steps_per_second": 4.767,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5554932355880737,
      "learning_rate": 7e-05,
      "loss": 0.1169,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.5160915851593018,
      "eval_runtime": 12.3934,
      "eval_samples_per_second": 302.58,
      "eval_steps_per_second": 4.761,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.13707447052001953,
      "learning_rate": 7e-05,
      "loss": 0.1008,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.47378286719322205,
      "eval_runtime": 12.5966,
      "eval_samples_per_second": 297.7,
      "eval_steps_per_second": 4.684,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0035035917535424232,
      "learning_rate": 7e-05,
      "loss": 0.0829,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.5170830488204956,
      "eval_runtime": 12.2958,
      "eval_samples_per_second": 304.981,
      "eval_steps_per_second": 4.798,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.27244240045547485,
      "learning_rate": 7e-05,
      "loss": 0.0779,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.5467396378517151,
      "eval_runtime": 12.2622,
      "eval_samples_per_second": 305.818,
      "eval_steps_per_second": 4.812,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.210299491882324,
      "learning_rate": 7e-05,
      "loss": 0.0698,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8853333333333333,
      "eval_loss": 0.5364003777503967,
      "eval_runtime": 12.4356,
      "eval_samples_per_second": 301.554,
      "eval_steps_per_second": 4.744,
      "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
}