trainer_state.json

{
  "best_metric": 0.9090666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0544/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": 42.7746696472168,
      "learning_rate": 0.000288581929876693,
      "loss": 1.4855,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8362666666666667,
      "eval_loss": 0.505395233631134,
      "eval_runtime": 7.3766,
      "eval_samples_per_second": 508.361,
      "eval_steps_per_second": 7.998,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 93.85247802734375,
      "learning_rate": 0.00025606601717798207,
      "loss": 0.3839,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.38962969183921814,
      "eval_runtime": 7.051,
      "eval_samples_per_second": 531.839,
      "eval_steps_per_second": 8.368,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 66.1702651977539,
      "learning_rate": 0.00020740251485476345,
      "loss": 0.1959,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8874666666666666,
      "eval_loss": 0.36873582005500793,
      "eval_runtime": 7.059,
      "eval_samples_per_second": 531.236,
      "eval_steps_per_second": 8.358,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 37.449398040771484,
      "learning_rate": 0.00015,
      "loss": 0.1035,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8906666666666667,
      "eval_loss": 0.3810472786426544,
      "eval_runtime": 7.077,
      "eval_samples_per_second": 529.882,
      "eval_steps_per_second": 8.337,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 67.34947967529297,
      "learning_rate": 9.259748514523653e-05,
      "loss": 0.0549,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9045333333333333,
      "eval_loss": 0.36172303557395935,
      "eval_runtime": 7.0237,
      "eval_samples_per_second": 533.907,
      "eval_steps_per_second": 8.4,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 44.025360107421875,
      "learning_rate": 4.3933982822017876e-05,
      "loss": 0.0292,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.908,
      "eval_loss": 0.357878178358078,
      "eval_runtime": 7.0663,
      "eval_samples_per_second": 530.689,
      "eval_steps_per_second": 8.349,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 38.25475311279297,
      "learning_rate": 1.1418070123306989e-05,
      "loss": 0.0163,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9074666666666666,
      "eval_loss": 0.3596632778644562,
      "eval_runtime": 7.063,
      "eval_samples_per_second": 530.937,
      "eval_steps_per_second": 8.353,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 76.10264587402344,
      "learning_rate": 0.0,
      "loss": 0.0128,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.3659333884716034,
      "eval_runtime": 7.0994,
      "eval_samples_per_second": 528.213,
      "eval_steps_per_second": 8.311,
      "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}