trainer_state.json

{
  "best_metric": 0.8834666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0891/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 5e-05,
      "loss": 1.7887,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8122666666666667,
      "eval_loss": 0.7417166829109192,
      "eval_runtime": 6.5128,
      "eval_samples_per_second": 575.785,
      "eval_steps_per_second": 9.059,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.5376,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8458666666666667,
      "eval_loss": 0.5409099459648132,
      "eval_runtime": 7.3353,
      "eval_samples_per_second": 511.225,
      "eval_steps_per_second": 8.043,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.3469,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.46692681312561035,
      "eval_runtime": 6.9679,
      "eval_samples_per_second": 538.181,
      "eval_steps_per_second": 8.467,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.19952133297920227,
      "learning_rate": 5e-05,
      "loss": 0.2407,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.44670984148979187,
      "eval_runtime": 6.581,
      "eval_samples_per_second": 569.824,
      "eval_steps_per_second": 8.965,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.0021045207977295,
      "learning_rate": 5e-05,
      "loss": 0.178,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.42581433057785034,
      "eval_runtime": 8.9545,
      "eval_samples_per_second": 418.783,
      "eval_steps_per_second": 6.589,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}