trainer_state.json

{
  "best_metric": 0.9061333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0333/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": 2.940774440765381,
      "learning_rate": 7.875e-05,
      "loss": 1.4508,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8050666666666667,
      "eval_loss": 0.7225078344345093,
      "eval_runtime": 5.9646,
      "eval_samples_per_second": 628.712,
      "eval_steps_per_second": 9.892,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9606209993362427,
      "learning_rate": 6.75e-05,
      "loss": 0.414,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.46382203698158264,
      "eval_runtime": 5.7649,
      "eval_samples_per_second": 650.487,
      "eval_steps_per_second": 10.234,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 5.628378378378378e-05,
      "loss": 0.2031,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8816,
      "eval_loss": 0.42458176612854004,
      "eval_runtime": 5.7916,
      "eval_samples_per_second": 647.491,
      "eval_steps_per_second": 10.187,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.32246658205986023,
      "learning_rate": 4.503378378378379e-05,
      "loss": 0.0947,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.4234279692173004,
      "eval_runtime": 5.7491,
      "eval_samples_per_second": 652.274,
      "eval_steps_per_second": 10.262,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07146262377500534,
      "learning_rate": 3.3783783783783784e-05,
      "loss": 0.0448,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.4155102074146271,
      "eval_runtime": 5.6773,
      "eval_samples_per_second": 660.527,
      "eval_steps_per_second": 10.392,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6648396849632263,
      "learning_rate": 2.2533783783783782e-05,
      "loss": 0.0142,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8981333333333333,
      "eval_loss": 0.40306609869003296,
      "eval_runtime": 5.9514,
      "eval_samples_per_second": 630.1,
      "eval_steps_per_second": 9.914,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.25383174419403076,
      "learning_rate": 1.1283783783783785e-05,
      "loss": 0.0055,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9048,
      "eval_loss": 0.39292892813682556,
      "eval_runtime": 5.7533,
      "eval_samples_per_second": 651.796,
      "eval_steps_per_second": 10.255,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.04514557123184204,
      "learning_rate": 3.3783783783783786e-08,
      "loss": 0.003,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9061333333333333,
      "eval_loss": 0.38685131072998047,
      "eval_runtime": 5.7713,
      "eval_samples_per_second": 649.762,
      "eval_steps_per_second": 10.223,
      "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
}