trainer_state.json

{
  "best_metric": 0.952,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0681/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": 0.28916457295417786,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 0.8285,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9328,
      "eval_loss": 0.24862666428089142,
      "eval_runtime": 3.9746,
      "eval_samples_per_second": 943.488,
      "eval_steps_per_second": 14.844,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.1916744709014893,
      "learning_rate": 4.267766952966369e-05,
      "loss": 0.1185,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9402666666666667,
      "eval_loss": 0.20952874422073364,
      "eval_runtime": 3.9741,
      "eval_samples_per_second": 943.613,
      "eval_steps_per_second": 14.846,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.403613328933716,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.0399,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9410666666666667,
      "eval_loss": 0.20657961070537567,
      "eval_runtime": 3.9402,
      "eval_samples_per_second": 951.729,
      "eval_steps_per_second": 14.974,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.01397969014942646,
      "learning_rate": 2.5e-05,
      "loss": 0.011,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.948,
      "eval_loss": 0.20767022669315338,
      "eval_runtime": 3.9317,
      "eval_samples_per_second": 953.793,
      "eval_steps_per_second": 15.006,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03432179242372513,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.0043,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9501333333333334,
      "eval_loss": 0.20706233382225037,
      "eval_runtime": 3.9431,
      "eval_samples_per_second": 951.037,
      "eval_steps_per_second": 14.963,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.1298101544380188,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.0022,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9517333333333333,
      "eval_loss": 0.2100895196199417,
      "eval_runtime": 3.9254,
      "eval_samples_per_second": 955.31,
      "eval_steps_per_second": 15.03,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 1.91430998468726e-06,
      "loss": 0.0016,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9517333333333333,
      "eval_loss": 0.21217672526836395,
      "eval_runtime": 3.9597,
      "eval_samples_per_second": 947.048,
      "eval_steps_per_second": 14.9,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0285886712372303,
      "learning_rate": 1.7383661915504334e-11,
      "loss": 0.0015,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.952,
      "eval_loss": 0.21255628764629364,
      "eval_runtime": 3.9879,
      "eval_samples_per_second": 940.336,
      "eval_steps_per_second": 14.795,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}