trainer_state.json

{
  "best_metric": 0.9378666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0775/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": 0.18666143715381622,
      "learning_rate": 9e-05,
      "loss": 0.6756,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9269333333333334,
      "eval_loss": 0.264104962348938,
      "eval_runtime": 11.4527,
      "eval_samples_per_second": 327.433,
      "eval_steps_per_second": 5.152,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.953794002532959,
      "learning_rate": 9e-05,
      "loss": 0.1538,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9301333333333334,
      "eval_loss": 0.24698102474212646,
      "eval_runtime": 11.4624,
      "eval_samples_per_second": 327.157,
      "eval_steps_per_second": 5.147,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.31670570373535156,
      "learning_rate": 9e-05,
      "loss": 0.0828,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9232,
      "eval_loss": 0.29440879821777344,
      "eval_runtime": 11.9709,
      "eval_samples_per_second": 313.26,
      "eval_steps_per_second": 4.929,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.021918971091508865,
      "learning_rate": 9e-05,
      "loss": 0.056,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9181333333333334,
      "eval_loss": 0.3319683074951172,
      "eval_runtime": 11.2961,
      "eval_samples_per_second": 331.974,
      "eval_steps_per_second": 5.223,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.012064099311828613,
      "learning_rate": 9e-05,
      "loss": 0.0462,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9274666666666667,
      "eval_loss": 0.3026920557022095,
      "eval_runtime": 11.4155,
      "eval_samples_per_second": 328.502,
      "eval_steps_per_second": 5.168,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.009431472048163414,
      "learning_rate": 9e-05,
      "loss": 0.033,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9378666666666666,
      "eval_loss": 0.2842191755771637,
      "eval_runtime": 11.4935,
      "eval_samples_per_second": 326.272,
      "eval_steps_per_second": 5.133,
      "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
}