trainer_state.json

{
  "best_metric": 0.8762666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0585/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": 6.127627627627627e-05,
      "loss": 1.7209,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7685333333333333,
      "eval_loss": 0.8285788297653198,
      "eval_runtime": 6.7659,
      "eval_samples_per_second": 554.254,
      "eval_steps_per_second": 8.72,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.505614757537842,
      "learning_rate": 5.252627627627627e-05,
      "loss": 0.6033,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8176,
      "eval_loss": 0.6332163214683533,
      "eval_runtime": 6.5959,
      "eval_samples_per_second": 568.536,
      "eval_steps_per_second": 8.945,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 4.3802552552552546e-05,
      "loss": 0.3576,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8456,
      "eval_loss": 0.534168541431427,
      "eval_runtime": 6.7571,
      "eval_samples_per_second": 554.969,
      "eval_steps_per_second": 8.732,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.987259864807129,
      "learning_rate": 3.5052552552552544e-05,
      "loss": 0.2152,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.5086409449577332,
      "eval_runtime": 6.6428,
      "eval_samples_per_second": 564.524,
      "eval_steps_per_second": 8.882,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7100839614868164,
      "learning_rate": 2.630255255255255e-05,
      "loss": 0.1268,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.48642468452453613,
      "eval_runtime": 6.6085,
      "eval_samples_per_second": 567.451,
      "eval_steps_per_second": 8.928,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.5768918991088867,
      "learning_rate": 1.7552552552552552e-05,
      "loss": 0.0731,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.864,
      "eval_loss": 0.49057745933532715,
      "eval_runtime": 7.4284,
      "eval_samples_per_second": 504.818,
      "eval_steps_per_second": 7.942,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.6284244656562805,
      "learning_rate": 8.802552552552553e-06,
      "loss": 0.0342,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.49280548095703125,
      "eval_runtime": 7.0901,
      "eval_samples_per_second": 528.907,
      "eval_steps_per_second": 8.321,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.46075132489204407,
      "learning_rate": 5.255255255255255e-08,
      "loss": 0.0176,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.48565638065338135,
      "eval_runtime": 6.5759,
      "eval_samples_per_second": 570.265,
      "eval_steps_per_second": 8.972,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}