trainer_state.json

{
  "best_metric": 0.5378666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0636/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": 0.0004850126564054637,
      "loss": 3.238,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.23173333333333335,
      "eval_loss": 2.8766443729400635,
      "eval_runtime": 6.9769,
      "eval_samples_per_second": 537.489,
      "eval_steps_per_second": 8.456,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00044184758936602033,
      "loss": 2.6907,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.29573333333333335,
      "eval_loss": 2.5490224361419678,
      "eval_runtime": 6.3165,
      "eval_samples_per_second": 593.686,
      "eval_steps_per_second": 9.341,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 12.459184646606445,
      "learning_rate": 0.0003754536287075859,
      "loss": 2.3365,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3621333333333333,
      "eval_loss": 2.3414154052734375,
      "eval_runtime": 6.3405,
      "eval_samples_per_second": 591.436,
      "eval_steps_per_second": 9.305,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00029418606892278544,
      "loss": 2.0399,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4136,
      "eval_loss": 2.1165003776550293,
      "eval_runtime": 6.3939,
      "eval_samples_per_second": 586.495,
      "eval_steps_per_second": 9.228,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 12.963085174560547,
      "learning_rate": 0.0002073624094068776,
      "loss": 1.711,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.47386666666666666,
      "eval_loss": 1.8930609226226807,
      "eval_runtime": 6.6654,
      "eval_samples_per_second": 562.604,
      "eval_steps_per_second": 8.852,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 20.189332962036133,
      "learning_rate": 0.00012568147258031897,
      "loss": 1.3432,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4965333333333333,
      "eval_loss": 1.821900725364685,
      "eval_runtime": 6.7781,
      "eval_samples_per_second": 553.249,
      "eval_steps_per_second": 8.704,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.813217639923096,
      "learning_rate": 5.899518490663266e-05,
      "loss": 0.9298,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5301333333333333,
      "eval_loss": 1.7683234214782715,
      "eval_runtime": 6.8334,
      "eval_samples_per_second": 548.774,
      "eval_steps_per_second": 8.634,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.188009738922119,
      "learning_rate": 1.534689686405272e-05,
      "loss": 0.5906,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5378666666666667,
      "eval_loss": 1.7697612047195435,
      "eval_runtime": 6.4443,
      "eval_samples_per_second": 581.906,
      "eval_steps_per_second": 9.155,
      "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
}