trainer_state.json

{
  "best_metric": 0.8536,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0665/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": 2.4301,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6253333333333333,
      "eval_loss": 1.2593867778778076,
      "eval_runtime": 12.0424,
      "eval_samples_per_second": 311.4,
      "eval_steps_per_second": 4.899,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5878950357437134,
      "learning_rate": 0.0002251126126126126,
      "loss": 0.8345,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7792,
      "eval_loss": 0.7301507592201233,
      "eval_runtime": 11.7407,
      "eval_samples_per_second": 319.401,
      "eval_steps_per_second": 5.025,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7901737093925476,
      "learning_rate": 0.00018761261261261258,
      "loss": 0.402,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7957333333333333,
      "eval_loss": 0.6763091087341309,
      "eval_runtime": 11.9066,
      "eval_samples_per_second": 314.952,
      "eval_steps_per_second": 4.955,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.5660219192504883,
      "learning_rate": 0.0001501126126126126,
      "loss": 0.1903,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8136,
      "eval_loss": 0.6362057328224182,
      "eval_runtime": 12.2823,
      "eval_samples_per_second": 305.318,
      "eval_steps_per_second": 4.804,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.38660183548927307,
      "learning_rate": 0.0001126126126126126,
      "loss": 0.0819,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8144,
      "eval_loss": 0.680377185344696,
      "eval_runtime": 11.999,
      "eval_samples_per_second": 312.525,
      "eval_steps_per_second": 4.917,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6845181584358215,
      "learning_rate": 7.51126126126126e-05,
      "loss": 0.0285,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8328,
      "eval_loss": 0.6544612050056458,
      "eval_runtime": 12.2037,
      "eval_samples_per_second": 307.284,
      "eval_steps_per_second": 4.835,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.17573580145835876,
      "learning_rate": 3.7612612612612614e-05,
      "loss": 0.0057,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8506666666666667,
      "eval_loss": 0.6123651266098022,
      "eval_runtime": 11.996,
      "eval_samples_per_second": 312.604,
      "eval_steps_per_second": 4.918,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.04138314351439476,
      "learning_rate": 1.1261261261261261e-07,
      "loss": 0.0016,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8536,
      "eval_loss": 0.6042613387107849,
      "eval_runtime": 12.035,
      "eval_samples_per_second": 311.591,
      "eval_steps_per_second": 4.902,
      "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
}