trainer_state.json

{
  "best_metric": 0.8869333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0569/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": 2.9100759384327822e-05,
      "loss": 2.3094,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7541333333333333,
      "eval_loss": 1.1307908296585083,
      "eval_runtime": 7.5264,
      "eval_samples_per_second": 498.245,
      "eval_steps_per_second": 7.839,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.6510855361961218e-05,
      "loss": 0.7601,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.6600532531738281,
      "eval_runtime": 6.66,
      "eval_samples_per_second": 563.063,
      "eval_steps_per_second": 8.859,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.9793548583984375,
      "learning_rate": 2.2527217722455156e-05,
      "loss": 0.4153,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8498666666666667,
      "eval_loss": 0.5441000461578369,
      "eval_runtime": 6.6394,
      "eval_samples_per_second": 564.809,
      "eval_steps_per_second": 8.886,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 62.02598571777344,
      "learning_rate": 1.763568653571882e-05,
      "loss": 0.2593,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4715833365917206,
      "eval_runtime": 6.6238,
      "eval_samples_per_second": 566.139,
      "eval_steps_per_second": 8.907,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 19.104129791259766,
      "learning_rate": 1.2426252654184347e-05,
      "loss": 0.1623,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.44464564323425293,
      "eval_runtime": 7.0662,
      "eval_samples_per_second": 530.694,
      "eval_steps_per_second": 8.35,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.8952959775924683,
      "learning_rate": 7.5272506870206094e-06,
      "loss": 0.1023,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.4369792342185974,
      "eval_runtime": 7.1105,
      "eval_samples_per_second": 527.391,
      "eval_steps_per_second": 8.298,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.37690165638923645,
      "learning_rate": 3.5295725730362483e-06,
      "loss": 0.0658,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.4345799386501312,
      "eval_runtime": 6.6663,
      "eval_samples_per_second": 562.53,
      "eval_steps_per_second": 8.85,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.294790267944336,
      "learning_rate": 9.153972922231291e-07,
      "loss": 0.0456,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8869333333333334,
      "eval_loss": 0.42035502195358276,
      "eval_runtime": 6.6418,
      "eval_samples_per_second": 564.602,
      "eval_steps_per_second": 8.883,
      "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
}