trainer_state.json

{
  "best_metric": 0.5773333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0331/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 0.0004850126564054637,
      "loss": 3.1956,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22773333333333334,
      "eval_loss": 2.8404667377471924,
      "eval_runtime": 7.6177,
      "eval_samples_per_second": 492.277,
      "eval_steps_per_second": 7.745,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00044184758936602033,
      "loss": 2.553,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.34346666666666664,
      "eval_loss": 2.352484941482544,
      "eval_runtime": 6.9285,
      "eval_samples_per_second": 541.242,
      "eval_steps_per_second": 8.516,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.451933860778809,
      "learning_rate": 0.0003754536287075859,
      "loss": 2.1182,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.42586666666666667,
      "eval_loss": 2.0866518020629883,
      "eval_runtime": 6.827,
      "eval_samples_per_second": 549.287,
      "eval_steps_per_second": 8.642,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 12.854957580566406,
      "learning_rate": 0.000293928108928647,
      "loss": 1.7619,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4816,
      "eval_loss": 1.8115888833999634,
      "eval_runtime": 7.3376,
      "eval_samples_per_second": 511.069,
      "eval_steps_per_second": 8.041,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.6629254817962646,
      "learning_rate": 0.00020710421090307243,
      "loss": 1.4225,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5178666666666667,
      "eval_loss": 1.7153759002685547,
      "eval_runtime": 8.0737,
      "eval_samples_per_second": 464.47,
      "eval_steps_per_second": 7.308,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 13.831657409667969,
      "learning_rate": 0.00012545417811701015,
      "loss": 1.0426,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5544,
      "eval_loss": 1.5989588499069214,
      "eval_runtime": 6.9124,
      "eval_samples_per_second": 542.5,
      "eval_steps_per_second": 8.535,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.8461707830429077,
      "learning_rate": 5.882620955060414e-05,
      "loss": 0.6771,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5738666666666666,
      "eval_loss": 1.5768643617630005,
      "eval_runtime": 6.8549,
      "eval_samples_per_second": 547.057,
      "eval_steps_per_second": 8.607,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 8.53959846496582,
      "learning_rate": 1.5256621537052151e-05,
      "loss": 0.4006,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5773333333333334,
      "eval_loss": 1.5561692714691162,
      "eval_runtime": 6.8055,
      "eval_samples_per_second": 551.028,
      "eval_steps_per_second": 8.67,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}