trainer_state.json

{
  "best_metric": 0.9184,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0164/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": 1.6117228269577026,
      "learning_rate": 0.0004375,
      "loss": 0.788,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8178666666666666,
      "eval_loss": 0.619030773639679,
      "eval_runtime": 7.872,
      "eval_samples_per_second": 476.373,
      "eval_steps_per_second": 7.495,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3596571683883667,
      "learning_rate": 0.000375,
      "loss": 0.3877,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8485333333333334,
      "eval_loss": 0.5202609300613403,
      "eval_runtime": 6.2911,
      "eval_samples_per_second": 596.077,
      "eval_steps_per_second": 9.378,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.02056737430393696,
      "learning_rate": 0.0003125,
      "loss": 0.235,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.4314245581626892,
      "eval_runtime": 6.4142,
      "eval_samples_per_second": 584.639,
      "eval_steps_per_second": 9.198,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.008273312821984291,
      "learning_rate": 0.00025,
      "loss": 0.1496,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8730666666666667,
      "eval_loss": 0.4702135920524597,
      "eval_runtime": 5.9537,
      "eval_samples_per_second": 629.863,
      "eval_steps_per_second": 9.91,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.20841361582279205,
      "learning_rate": 0.0001875,
      "loss": 0.0752,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.433009535074234,
      "eval_runtime": 7.5515,
      "eval_samples_per_second": 496.589,
      "eval_steps_per_second": 7.813,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.008517544716596603,
      "learning_rate": 0.000125,
      "loss": 0.0339,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8989333333333334,
      "eval_loss": 0.41774675250053406,
      "eval_runtime": 6.3262,
      "eval_samples_per_second": 592.773,
      "eval_steps_per_second": 9.326,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 6.26876876876877e-05,
      "loss": 0.0117,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9149333333333334,
      "eval_loss": 0.3844277262687683,
      "eval_runtime": 6.3136,
      "eval_samples_per_second": 593.953,
      "eval_steps_per_second": 9.345,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0019493737490847707,
      "learning_rate": 1.876876876876877e-07,
      "loss": 0.0024,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9184,
      "eval_loss": 0.3688521683216095,
      "eval_runtime": 5.8345,
      "eval_samples_per_second": 642.731,
      "eval_steps_per_second": 10.112,
      "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
}