trainer_state.json

{
  "best_metric": 0.9362666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0779/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": 2.7254366874694824,
      "learning_rate": 7e-05,
      "loss": 0.6588,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9197333333333333,
      "eval_loss": 0.2626796364784241,
      "eval_runtime": 12.376,
      "eval_samples_per_second": 303.006,
      "eval_steps_per_second": 4.767,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3961549699306488,
      "learning_rate": 7e-05,
      "loss": 0.1337,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9304,
      "eval_loss": 0.23372413218021393,
      "eval_runtime": 11.8194,
      "eval_samples_per_second": 317.274,
      "eval_steps_per_second": 4.992,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.0728,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9296,
      "eval_loss": 0.25086480379104614,
      "eval_runtime": 12.1103,
      "eval_samples_per_second": 309.655,
      "eval_steps_per_second": 4.872,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.02090725302696228,
      "learning_rate": 7e-05,
      "loss": 0.0414,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9352,
      "eval_loss": 0.24080999195575714,
      "eval_runtime": 13.8337,
      "eval_samples_per_second": 271.076,
      "eval_steps_per_second": 4.265,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.008636542595922947,
      "learning_rate": 7e-05,
      "loss": 0.0284,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9330666666666667,
      "eval_loss": 0.26541298627853394,
      "eval_runtime": 12.326,
      "eval_samples_per_second": 304.234,
      "eval_steps_per_second": 4.787,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.006616988684982061,
      "learning_rate": 7e-05,
      "loss": 0.0224,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9362666666666667,
      "eval_loss": 0.28036877512931824,
      "eval_runtime": 12.2794,
      "eval_samples_per_second": 305.39,
      "eval_steps_per_second": 4.805,
      "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
}