trainer_state.json

{
  "best_metric": 0.8589333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0124/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": 3e-05,
      "loss": 2.2696,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7104,
      "eval_loss": 1.196899175643921,
      "eval_runtime": 6.2587,
      "eval_samples_per_second": 599.168,
      "eval_steps_per_second": 9.427,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 0.7545,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8077333333333333,
      "eval_loss": 0.6977005004882812,
      "eval_runtime": 5.9965,
      "eval_samples_per_second": 625.37,
      "eval_steps_per_second": 9.839,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.3362598419189453,
      "learning_rate": 3e-05,
      "loss": 0.4006,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8482666666666666,
      "eval_loss": 0.5300347208976746,
      "eval_runtime": 5.5385,
      "eval_samples_per_second": 677.073,
      "eval_steps_per_second": 10.653,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8317432403564453,
      "learning_rate": 3e-05,
      "loss": 0.2324,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.4805912673473358,
      "eval_runtime": 5.543,
      "eval_samples_per_second": 676.531,
      "eval_steps_per_second": 10.644,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 11.36332893371582,
      "learning_rate": 3e-05,
      "loss": 0.1348,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.5063788890838623,
      "eval_runtime": 5.6346,
      "eval_samples_per_second": 665.529,
      "eval_steps_per_second": 10.471,
      "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
}