trainer_state.json

{
  "best_metric": 0.9173333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0689/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": 14.203742980957031,
      "learning_rate": 8e-05,
      "loss": 1.4913,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8261333333333334,
      "eval_loss": 0.6348932385444641,
      "eval_runtime": 5.965,
      "eval_samples_per_second": 628.67,
      "eval_steps_per_second": 9.891,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.3171796798706055,
      "learning_rate": 6e-05,
      "loss": 0.4191,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.872,
      "eval_loss": 0.441319078207016,
      "eval_runtime": 5.9714,
      "eval_samples_per_second": 627.988,
      "eval_steps_per_second": 9.88,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.115429401397705,
      "learning_rate": 4e-05,
      "loss": 0.2034,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8912,
      "eval_loss": 0.3866879642009735,
      "eval_runtime": 6.0884,
      "eval_samples_per_second": 615.927,
      "eval_steps_per_second": 9.691,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8428359627723694,
      "learning_rate": 2e-05,
      "loss": 0.0946,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9053333333333333,
      "eval_loss": 0.3259713649749756,
      "eval_runtime": 5.881,
      "eval_samples_per_second": 637.645,
      "eval_steps_per_second": 10.032,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7991333603858948,
      "learning_rate": 0.0,
      "loss": 0.0386,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.30637913942337036,
      "eval_runtime": 5.9226,
      "eval_samples_per_second": 633.173,
      "eval_steps_per_second": 9.962,
      "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
}