trainer_state.json

{
  "best_metric": 0.8536,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0207/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": 0.00026261261261261256,
      "loss": 2.72,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5266666666666666,
      "eval_loss": 1.6845567226409912,
      "eval_runtime": 6.4011,
      "eval_samples_per_second": 585.837,
      "eval_steps_per_second": 9.217,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0002252252252252252,
      "loss": 1.123,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7306666666666667,
      "eval_loss": 0.8867805600166321,
      "eval_runtime": 5.6715,
      "eval_samples_per_second": 661.195,
      "eval_steps_per_second": 10.403,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 6.955826282501221,
      "learning_rate": 0.00018772522522522518,
      "loss": 0.5725,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7842666666666667,
      "eval_loss": 0.7251701951026917,
      "eval_runtime": 5.6857,
      "eval_samples_per_second": 659.55,
      "eval_steps_per_second": 10.377,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.09944434463977814,
      "learning_rate": 0.0001502252252252252,
      "loss": 0.291,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8106666666666666,
      "eval_loss": 0.6697235703468323,
      "eval_runtime": 5.7527,
      "eval_samples_per_second": 651.863,
      "eval_steps_per_second": 10.256,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9137321710586548,
      "learning_rate": 0.00011272522522522522,
      "loss": 0.1408,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8168,
      "eval_loss": 0.6312540173530579,
      "eval_runtime": 5.6986,
      "eval_samples_per_second": 658.054,
      "eval_steps_per_second": 10.353,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.054368551820516586,
      "learning_rate": 7.522522522522523e-05,
      "loss": 0.0665,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8394666666666667,
      "eval_loss": 0.5909776091575623,
      "eval_runtime": 5.801,
      "eval_samples_per_second": 646.436,
      "eval_steps_per_second": 10.171,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.2923530638217926,
      "learning_rate": 3.7725225225225225e-05,
      "loss": 0.0235,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5728355646133423,
      "eval_runtime": 5.675,
      "eval_samples_per_second": 660.795,
      "eval_steps_per_second": 10.397,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.28572508692741394,
      "learning_rate": 2.2522522522522522e-07,
      "loss": 0.0069,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8536,
      "eval_loss": 0.55720454454422,
      "eval_runtime": 6.8328,
      "eval_samples_per_second": 548.82,
      "eval_steps_per_second": 8.635,
      "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
}