trainer_state.json

{
  "best_metric": 0.9437333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0919/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": 1.4914759397506714,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 0.8682,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9221333333333334,
      "eval_loss": 0.26893118023872375,
      "eval_runtime": 4.6675,
      "eval_samples_per_second": 803.431,
      "eval_steps_per_second": 12.641,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.751140117645264,
      "learning_rate": 4.267766952966369e-05,
      "loss": 0.1589,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9344,
      "eval_loss": 0.2224694937467575,
      "eval_runtime": 4.655,
      "eval_samples_per_second": 805.583,
      "eval_steps_per_second": 12.675,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.054097067564725876,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.0717,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9357333333333333,
      "eval_loss": 0.23097339272499084,
      "eval_runtime": 4.5657,
      "eval_samples_per_second": 821.348,
      "eval_steps_per_second": 12.923,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.057016536593437195,
      "learning_rate": 2.5e-05,
      "loss": 0.0331,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.23120594024658203,
      "eval_runtime": 4.6004,
      "eval_samples_per_second": 815.154,
      "eval_steps_per_second": 12.825,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.01529560424387455,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.0135,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9416,
      "eval_loss": 0.22757594287395477,
      "eval_runtime": 4.5475,
      "eval_samples_per_second": 824.624,
      "eval_steps_per_second": 12.974,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0630682185292244,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.0067,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9437333333333333,
      "eval_loss": 0.22726932168006897,
      "eval_runtime": 4.7516,
      "eval_samples_per_second": 789.206,
      "eval_steps_per_second": 12.417,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}