trainer_state.json

{
  "best_metric": 0.5290666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0458/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": 0.0004810825448943271,
      "loss": 3.2394,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.21786666666666665,
      "eval_loss": 2.8672330379486084,
      "eval_runtime": 4.4172,
      "eval_samples_per_second": 848.957,
      "eval_steps_per_second": 13.357,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0004271931404387096,
      "loss": 2.653,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3048,
      "eval_loss": 2.510408878326416,
      "eval_runtime": 5.0688,
      "eval_samples_per_second": 739.817,
      "eval_steps_per_second": 11.64,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.560858726501465,
      "learning_rate": 0.0003462153462094779,
      "loss": 2.2417,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.37173333333333336,
      "eval_loss": 2.2264952659606934,
      "eval_runtime": 4.3739,
      "eval_samples_per_second": 857.365,
      "eval_steps_per_second": 13.489,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00025088445554620134,
      "loss": 1.9285,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.42213333333333336,
      "eval_loss": 2.019205331802368,
      "eval_runtime": 5.3116,
      "eval_samples_per_second": 706.008,
      "eval_steps_per_second": 11.108,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 13.726110458374023,
      "learning_rate": 0.00015514687100427082,
      "loss": 1.5806,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.4885333333333333,
      "eval_loss": 1.8193490505218506,
      "eval_runtime": 4.3055,
      "eval_samples_per_second": 870.988,
      "eval_steps_per_second": 13.704,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 15.254719734191895,
      "learning_rate": 7.384981550618446e-05,
      "loss": 1.1929,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5106666666666667,
      "eval_loss": 1.745268702507019,
      "eval_runtime": 4.3105,
      "eval_samples_per_second": 869.962,
      "eval_steps_per_second": 13.687,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.193294525146484,
      "learning_rate": 1.9370028791882848e-05,
      "loss": 0.8364,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5269333333333334,
      "eval_loss": 1.6878172159194946,
      "eval_runtime": 4.315,
      "eval_samples_per_second": 869.055,
      "eval_steps_per_second": 13.673,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 11.877822875976562,
      "learning_rate": 1.5645281219167639e-09,
      "loss": 0.6429,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5290666666666667,
      "eval_loss": 1.694595217704773,
      "eval_runtime": 4.3276,
      "eval_samples_per_second": 866.524,
      "eval_steps_per_second": 13.633,
      "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
}