trainer_state.json

{
  "best_metric": 0.8824,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0745/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": 2.1105,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7746666666666666,
      "eval_loss": 0.9829122424125671,
      "eval_runtime": 10.8239,
      "eval_samples_per_second": 346.455,
      "eval_steps_per_second": 5.451,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 2.6510855361961218e-05,
      "loss": 0.6663,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.5813502073287964,
      "eval_runtime": 10.6948,
      "eval_samples_per_second": 350.636,
      "eval_steps_per_second": 5.517,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.826977252960205,
      "learning_rate": 2.2527217722455156e-05,
      "loss": 0.3838,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.5004039406776428,
      "eval_runtime": 11.0475,
      "eval_samples_per_second": 339.443,
      "eval_steps_per_second": 5.341,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.728541374206543,
      "learning_rate": 1.763568653571882e-05,
      "loss": 0.251,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.4512368440628052,
      "eval_runtime": 11.2474,
      "eval_samples_per_second": 333.41,
      "eval_steps_per_second": 5.246,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5031620264053345,
      "learning_rate": 1.2426252654184347e-05,
      "loss": 0.1613,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.4258069097995758,
      "eval_runtime": 11.1203,
      "eval_samples_per_second": 337.22,
      "eval_steps_per_second": 5.306,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 14.28777027130127,
      "learning_rate": 7.5272506870206094e-06,
      "loss": 0.1053,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.41303834319114685,
      "eval_runtime": 10.9444,
      "eval_samples_per_second": 342.642,
      "eval_steps_per_second": 5.391,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.20337265729904175,
      "learning_rate": 3.5295725730362483e-06,
      "loss": 0.0657,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8816,
      "eval_loss": 0.4092414677143097,
      "eval_runtime": 11.1338,
      "eval_samples_per_second": 336.813,
      "eval_steps_per_second": 5.299,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.7813453078269958,
      "learning_rate": 9.153972922231291e-07,
      "loss": 0.0464,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.4122046232223511,
      "eval_runtime": 11.2688,
      "eval_samples_per_second": 332.778,
      "eval_steps_per_second": 5.236,
      "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
}