trainer_state.json

{
  "best_metric": 0.896,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0801/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": 8.730227815298347e-05,
      "loss": 1.5101,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8104,
      "eval_loss": 0.7066605687141418,
      "eval_runtime": 7.2235,
      "eval_samples_per_second": 519.142,
      "eval_steps_per_second": 8.168,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.953256608588366e-05,
      "loss": 0.4515,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8421333333333333,
      "eval_loss": 0.5405563712120056,
      "eval_runtime": 6.6037,
      "eval_samples_per_second": 567.865,
      "eval_steps_per_second": 8.934,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 80.70404815673828,
      "learning_rate": 6.758165316736547e-05,
      "loss": 0.2303,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.47567427158355713,
      "eval_runtime": 6.6163,
      "eval_samples_per_second": 566.779,
      "eval_steps_per_second": 8.917,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.574110984802246,
      "learning_rate": 5.2907059607156464e-05,
      "loss": 0.1108,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.4540869891643524,
      "eval_runtime": 6.7226,
      "eval_samples_per_second": 557.819,
      "eval_steps_per_second": 8.776,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.11755507439374924,
      "learning_rate": 3.727875796255304e-05,
      "loss": 0.05,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.4773838520050049,
      "eval_runtime": 6.6412,
      "eval_samples_per_second": 564.657,
      "eval_steps_per_second": 8.884,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.1838432401418686,
      "learning_rate": 2.258175206106183e-05,
      "loss": 0.0176,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8853333333333333,
      "eval_loss": 0.4518565237522125,
      "eval_runtime": 7.3627,
      "eval_samples_per_second": 509.321,
      "eval_steps_per_second": 8.013,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.07572843134403229,
      "learning_rate": 1.0588717719108745e-05,
      "loss": 0.0048,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8941333333333333,
      "eval_loss": 0.4358372986316681,
      "eval_runtime": 6.6675,
      "eval_samples_per_second": 562.43,
      "eval_steps_per_second": 8.849,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.03222636133432388,
      "learning_rate": 2.7461918766693875e-06,
      "loss": 0.0025,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.43342873454093933,
      "eval_runtime": 6.6065,
      "eval_samples_per_second": 567.624,
      "eval_steps_per_second": 8.931,
      "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
}