trainer_state.json

{
  "best_metric": 0.7408,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0066/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": 117.7275390625,
      "learning_rate": 2.7990381056766583e-05,
      "loss": 3.856,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.29973333333333335,
      "eval_loss": 3.7554378509521484,
      "eval_runtime": 8.0628,
      "eval_samples_per_second": 465.098,
      "eval_steps_per_second": 7.318,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 148.98995971679688,
      "learning_rate": 2.25e-05,
      "loss": 3.3112,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5146666666666667,
      "eval_loss": 2.753460645675659,
      "eval_runtime": 8.1258,
      "eval_samples_per_second": 461.494,
      "eval_steps_per_second": 7.261,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 164.59437561035156,
      "learning_rate": 1.5e-05,
      "loss": 2.0949,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6642666666666667,
      "eval_loss": 1.5616481304168701,
      "eval_runtime": 8.0688,
      "eval_samples_per_second": 464.75,
      "eval_steps_per_second": 7.312,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 181.8090057373047,
      "learning_rate": 7.500000000000004e-06,
      "loss": 1.3829,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.728,
      "eval_loss": 1.2409359216690063,
      "eval_runtime": 8.0118,
      "eval_samples_per_second": 468.061,
      "eval_steps_per_second": 7.364,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 248.22152709960938,
      "learning_rate": 2.0096189432334194e-06,
      "loss": 1.1785,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7341333333333333,
      "eval_loss": 1.1382898092269897,
      "eval_runtime": 8.0115,
      "eval_samples_per_second": 468.076,
      "eval_steps_per_second": 7.364,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 182.55575561523438,
      "learning_rate": 0.0,
      "loss": 1.1278,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7408,
      "eval_loss": 1.1101104021072388,
      "eval_runtime": 8.5334,
      "eval_samples_per_second": 439.452,
      "eval_steps_per_second": 6.914,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}