trainer_state.json

{
  "best_metric": 0.9530666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0151/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.5435671210289001,
      "learning_rate": 4.522542485937369e-05,
      "loss": 0.8791,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9314666666666667,
      "eval_loss": 0.2558072805404663,
      "eval_runtime": 6.7945,
      "eval_samples_per_second": 551.921,
      "eval_steps_per_second": 8.684,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.47342240810394287,
      "learning_rate": 3.272542485937369e-05,
      "loss": 0.1177,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9408,
      "eval_loss": 0.20322370529174805,
      "eval_runtime": 7.3371,
      "eval_samples_per_second": 511.103,
      "eval_steps_per_second": 8.041,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.634818971157074,
      "learning_rate": 1.7274575140626318e-05,
      "loss": 0.0366,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9501333333333334,
      "eval_loss": 0.17973017692565918,
      "eval_runtime": 6.645,
      "eval_samples_per_second": 564.332,
      "eval_steps_per_second": 8.879,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.03654947131872177,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.0127,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9528,
      "eval_loss": 0.17758025228977203,
      "eval_runtime": 6.5938,
      "eval_samples_per_second": 568.717,
      "eval_steps_per_second": 8.948,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.054814886301755905,
      "learning_rate": 0.0,
      "loss": 0.0068,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9530666666666666,
      "eval_loss": 0.17499805986881256,
      "eval_runtime": 7.3231,
      "eval_samples_per_second": 512.079,
      "eval_steps_per_second": 8.057,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}