trainer_state.json

{
  "best_metric": 0.8349333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0605/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": 29.653953552246094,
      "learning_rate": 0.0001,
      "loss": 1.5958,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7784,
      "eval_loss": 0.7681472897529602,
      "eval_runtime": 6.5636,
      "eval_samples_per_second": 571.337,
      "eval_steps_per_second": 8.989,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.5848,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8189333333333333,
      "eval_loss": 0.5933331847190857,
      "eval_runtime": 6.4585,
      "eval_samples_per_second": 580.626,
      "eval_steps_per_second": 9.135,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.3649,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8346666666666667,
      "eval_loss": 0.5294280648231506,
      "eval_runtime": 6.6021,
      "eval_samples_per_second": 568.004,
      "eval_steps_per_second": 8.937,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.2524,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8333333333333334,
      "eval_loss": 0.5485849976539612,
      "eval_runtime": 6.4809,
      "eval_samples_per_second": 578.62,
      "eval_steps_per_second": 9.104,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07457827776670456,
      "learning_rate": 0.0001,
      "loss": 0.1863,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8349333333333333,
      "eval_loss": 0.5679844617843628,
      "eval_runtime": 6.5348,
      "eval_samples_per_second": 573.855,
      "eval_steps_per_second": 9.029,
      "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
}