trainer_state.json

{
  "best_metric": 0.8981333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0946/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": 8.730227815298347e-05,
      "loss": 1.6508,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7970666666666667,
      "eval_loss": 0.7427321672439575,
      "eval_runtime": 7.4042,
      "eval_samples_per_second": 506.467,
      "eval_steps_per_second": 7.968,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.35150566697120667,
      "learning_rate": 7.95023019690139e-05,
      "loss": 0.5549,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8269333333333333,
      "eval_loss": 0.5848798155784607,
      "eval_runtime": 7.6395,
      "eval_samples_per_second": 490.872,
      "eval_steps_per_second": 7.723,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.8437705636024475,
      "learning_rate": 6.75408389678355e-05,
      "loss": 0.3397,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8544,
      "eval_loss": 0.47647079825401306,
      "eval_runtime": 7.4177,
      "eval_samples_per_second": 505.545,
      "eval_steps_per_second": 7.954,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.3044145107269287,
      "learning_rate": 5.286061811978304e-05,
      "loss": 0.2107,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.46863865852355957,
      "eval_runtime": 7.4454,
      "eval_samples_per_second": 503.666,
      "eval_steps_per_second": 7.924,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.49341055750846863,
      "learning_rate": 3.723229071611673e-05,
      "loss": 0.1213,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.4322805404663086,
      "eval_runtime": 7.5417,
      "eval_samples_per_second": 497.233,
      "eval_steps_per_second": 7.823,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.20676162838935852,
      "learning_rate": 2.2540863691266368e-05,
      "loss": 0.0633,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8882666666666666,
      "eval_loss": 0.4021604061126709,
      "eval_runtime": 7.9603,
      "eval_samples_per_second": 471.09,
      "eval_steps_per_second": 7.412,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.3300224840641022,
      "learning_rate": 1.055833996679956e-05,
      "loss": 0.0267,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8962666666666667,
      "eval_loss": 0.3826501667499542,
      "eval_runtime": 7.5109,
      "eval_samples_per_second": 499.273,
      "eval_steps_per_second": 7.855,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.04048081114888191,
      "learning_rate": 2.7299887471028354e-06,
      "loss": 0.0138,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8981333333333333,
      "eval_loss": 0.3784742057323456,
      "eval_runtime": 7.5327,
      "eval_samples_per_second": 497.831,
      "eval_steps_per_second": 7.833,
      "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
}