trainer_state.json

{
  "best_metric": 0.8301333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0110/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 185.49679565429688,
      "learning_rate": 2.5714285714285714e-05,
      "loss": 3.8393,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.42213333333333336,
      "eval_loss": 3.714442014694214,
      "eval_runtime": 14.849,
      "eval_samples_per_second": 252.542,
      "eval_steps_per_second": 3.973,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 215.71383666992188,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 3.2019,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6469333333333334,
      "eval_loss": 2.637028932571411,
      "eval_runtime": 14.6676,
      "eval_samples_per_second": 255.665,
      "eval_steps_per_second": 4.022,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 195.22622680664062,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 1.8859,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7573333333333333,
      "eval_loss": 1.3570529222488403,
      "eval_runtime": 14.4332,
      "eval_samples_per_second": 259.818,
      "eval_steps_per_second": 4.088,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 100.56300354003906,
      "learning_rate": 1.2857142857142857e-05,
      "loss": 1.0691,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8082666666666667,
      "eval_loss": 0.8815708160400391,
      "eval_runtime": 14.8621,
      "eval_samples_per_second": 252.32,
      "eval_steps_per_second": 3.97,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 112.47240447998047,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.7843,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8162666666666667,
      "eval_loss": 0.7410248517990112,
      "eval_runtime": 14.0343,
      "eval_samples_per_second": 267.203,
      "eval_steps_per_second": 4.204,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 161.8098907470703,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 0.6687,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8301333333333333,
      "eval_loss": 0.6745153665542603,
      "eval_runtime": 14.0312,
      "eval_samples_per_second": 267.261,
      "eval_steps_per_second": 4.205,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}