trainer_state.json

{
  "best_metric": 0.9506666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0593/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": 21.177248001098633,
      "learning_rate": 9.504844339512095e-05,
      "loss": 0.6172,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9072,
      "eval_loss": 0.31727510690689087,
      "eval_runtime": 11.9449,
      "eval_samples_per_second": 313.942,
      "eval_steps_per_second": 4.939,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.06636176258325577,
      "learning_rate": 8.117449009293668e-05,
      "loss": 0.145,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9376,
      "eval_loss": 0.2289436161518097,
      "eval_runtime": 11.0662,
      "eval_samples_per_second": 338.868,
      "eval_steps_per_second": 5.332,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0420868955552578,
      "learning_rate": 6.112604669781572e-05,
      "loss": 0.0644,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.932,
      "eval_loss": 0.251239150762558,
      "eval_runtime": 11.363,
      "eval_samples_per_second": 330.019,
      "eval_steps_per_second": 5.192,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.06610515713691711,
      "learning_rate": 3.887395330218429e-05,
      "loss": 0.028,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9450666666666667,
      "eval_loss": 0.20461751520633698,
      "eval_runtime": 11.2246,
      "eval_samples_per_second": 334.089,
      "eval_steps_per_second": 5.256,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.020192207768559456,
      "learning_rate": 1.8825509907063327e-05,
      "loss": 0.0089,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9498666666666666,
      "eval_loss": 0.20890891551971436,
      "eval_runtime": 11.4213,
      "eval_samples_per_second": 328.333,
      "eval_steps_per_second": 5.166,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.014118768274784088,
      "learning_rate": 4.951556604879048e-06,
      "loss": 0.0042,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9506666666666667,
      "eval_loss": 0.20060475170612335,
      "eval_runtime": 10.8007,
      "eval_samples_per_second": 347.199,
      "eval_steps_per_second": 5.463,
      "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
}