trainer_state.json

{
  "best_metric": 0.9429333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0224/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": 0.5630245804786682,
      "learning_rate": 4.522542485937369e-05,
      "loss": 0.857,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9301333333333334,
      "eval_loss": 0.2612086236476898,
      "eval_runtime": 5.5021,
      "eval_samples_per_second": 681.559,
      "eval_steps_per_second": 10.723,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.3631815910339355,
      "learning_rate": 3.272542485937369e-05,
      "loss": 0.1288,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9328,
      "eval_loss": 0.21966294944286346,
      "eval_runtime": 5.4648,
      "eval_samples_per_second": 686.21,
      "eval_steps_per_second": 10.796,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.06337760388851166,
      "learning_rate": 1.7274575140626318e-05,
      "loss": 0.0368,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.94,
      "eval_loss": 0.23024700582027435,
      "eval_runtime": 5.8086,
      "eval_samples_per_second": 645.596,
      "eval_steps_per_second": 10.157,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.8840432167053223,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.0112,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9410666666666667,
      "eval_loss": 0.23338843882083893,
      "eval_runtime": 5.6919,
      "eval_samples_per_second": 658.83,
      "eval_steps_per_second": 10.366,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.1127036064863205,
      "learning_rate": 0.0,
      "loss": 0.0065,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9429333333333333,
      "eval_loss": 0.23060543835163116,
      "eval_runtime": 5.5696,
      "eval_samples_per_second": 673.301,
      "eval_steps_per_second": 10.593,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}