trainer_state.json

{
  "best_metric": 0.8413333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0357/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": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.8296,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8224,
      "eval_loss": 0.5756388306617737,
      "eval_runtime": 5.8388,
      "eval_samples_per_second": 642.257,
      "eval_steps_per_second": 10.105,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 12.801283836364746,
      "learning_rate": 0.0001,
      "loss": 0.3085,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8274666666666667,
      "eval_loss": 0.6019585728645325,
      "eval_runtime": 5.737,
      "eval_samples_per_second": 653.653,
      "eval_steps_per_second": 10.284,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.29244908690452576,
      "learning_rate": 0.0001,
      "loss": 0.2038,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8309333333333333,
      "eval_loss": 0.6679965257644653,
      "eval_runtime": 5.6123,
      "eval_samples_per_second": 668.181,
      "eval_steps_per_second": 10.513,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.00837612058967352,
      "learning_rate": 0.0001,
      "loss": 0.1632,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8336,
      "eval_loss": 0.6244806051254272,
      "eval_runtime": 5.5961,
      "eval_samples_per_second": 670.112,
      "eval_steps_per_second": 10.543,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0035336606670171022,
      "learning_rate": 0.0001,
      "loss": 0.1397,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8242666666666667,
      "eval_loss": 0.6966049671173096,
      "eval_runtime": 6.4278,
      "eval_samples_per_second": 583.401,
      "eval_steps_per_second": 9.179,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 86.37367248535156,
      "learning_rate": 0.0001,
      "loss": 0.1194,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.6402702331542969,
      "eval_runtime": 5.8767,
      "eval_samples_per_second": 638.11,
      "eval_steps_per_second": 10.04,
      "step": 1998
    }
  ],
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}