trainer_state.json

{
  "best_metric": 0.8989333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0561/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": Infinity,
      "learning_rate": 0.0004810825448943271,
      "loss": 0.8697,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7944,
      "eval_loss": 0.6976803541183472,
      "eval_runtime": 13.432,
      "eval_samples_per_second": 279.184,
      "eval_steps_per_second": 4.392,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.42569294571876526,
      "learning_rate": 0.0004269850409335979,
      "loss": 0.5026,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8074666666666667,
      "eval_loss": 0.6719813942909241,
      "eval_runtime": 15.9253,
      "eval_samples_per_second": 235.475,
      "eval_steps_per_second": 3.705,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.033622656017541885,
      "learning_rate": 0.0003459431688641196,
      "loss": 0.3307,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8205333333333333,
      "eval_loss": 0.6323661208152771,
      "eval_runtime": 13.3414,
      "eval_samples_per_second": 281.081,
      "eval_steps_per_second": 4.422,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.06341785192489624,
      "learning_rate": 0.00025029481906207064,
      "loss": 0.1964,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8496,
      "eval_loss": 0.516577422618866,
      "eval_runtime": 13.4318,
      "eval_samples_per_second": 279.188,
      "eval_steps_per_second": 4.393,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.00801101140677929,
      "learning_rate": 0.00015460158573036287,
      "loss": 0.0973,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8669333333333333,
      "eval_loss": 0.5082181096076965,
      "eval_runtime": 13.352,
      "eval_samples_per_second": 280.857,
      "eval_steps_per_second": 4.419,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.02043599635362625,
      "learning_rate": 7.343189618242854e-05,
      "loss": 0.0313,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.4650340676307678,
      "eval_runtime": 13.2599,
      "eval_samples_per_second": 282.807,
      "eval_steps_per_second": 4.449,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.026404811069369316,
      "learning_rate": 1.9143099846872597e-05,
      "loss": 0.0091,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.42266568541526794,
      "eval_runtime": 13.3416,
      "eval_samples_per_second": 281.076,
      "eval_steps_per_second": 4.422,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.09865768998861313,
      "learning_rate": 1.7383661915504333e-10,
      "loss": 0.0036,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.4371086061000824,
      "eval_runtime": 13.4849,
      "eval_samples_per_second": 278.089,
      "eval_steps_per_second": 4.375,
      "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
}