trainer_state.json

{
  "best_metric": 0.86,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0957/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": Infinity,
      "learning_rate": 7e-05,
      "loss": 2.099,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7432,
      "eval_loss": 0.9111053347587585,
      "eval_runtime": 5.8627,
      "eval_samples_per_second": 639.634,
      "eval_steps_per_second": 10.064,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 23.389272689819336,
      "learning_rate": 7e-05,
      "loss": 0.6168,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8242666666666667,
      "eval_loss": 0.6114821434020996,
      "eval_runtime": 4.9059,
      "eval_samples_per_second": 764.388,
      "eval_steps_per_second": 12.026,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.3781,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5333465337753296,
      "eval_runtime": 6.6072,
      "eval_samples_per_second": 567.563,
      "eval_steps_per_second": 8.93,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.384999752044678,
      "learning_rate": 7e-05,
      "loss": 0.2584,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8538666666666667,
      "eval_loss": 0.4912506639957428,
      "eval_runtime": 5.7643,
      "eval_samples_per_second": 650.551,
      "eval_steps_per_second": 10.235,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.5232985019683838,
      "learning_rate": 7e-05,
      "loss": 0.1994,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.86,
      "eval_loss": 0.5012106895446777,
      "eval_runtime": 4.4495,
      "eval_samples_per_second": 842.795,
      "eval_steps_per_second": 13.26,
      "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
}