trainer_state.json

{
  "best_metric": 0.9074666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0601/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": 136.4054718017578,
      "learning_rate": 9e-05,
      "loss": 2.7234,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7805333333333333,
      "eval_loss": 0.855254054069519,
      "eval_runtime": 7.6775,
      "eval_samples_per_second": 488.442,
      "eval_steps_per_second": 7.685,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 116.6287841796875,
      "learning_rate": 9e-05,
      "loss": 0.6258,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.4291454553604126,
      "eval_runtime": 7.6658,
      "eval_samples_per_second": 489.184,
      "eval_steps_per_second": 7.697,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 228.27731323242188,
      "learning_rate": 9e-05,
      "loss": 0.381,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.37970420718193054,
      "eval_runtime": 7.6018,
      "eval_samples_per_second": 493.307,
      "eval_steps_per_second": 7.761,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 104.64036560058594,
      "learning_rate": 9e-05,
      "loss": 0.2677,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8981333333333333,
      "eval_loss": 0.34468144178390503,
      "eval_runtime": 7.6258,
      "eval_samples_per_second": 491.754,
      "eval_steps_per_second": 7.737,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 110.99284362792969,
      "learning_rate": 9e-05,
      "loss": 0.1912,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8933333333333333,
      "eval_loss": 0.3526991307735443,
      "eval_runtime": 7.6427,
      "eval_samples_per_second": 490.665,
      "eval_steps_per_second": 7.72,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 98.84141540527344,
      "learning_rate": 9e-05,
      "loss": 0.1384,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.34339040517807007,
      "eval_runtime": 7.6664,
      "eval_samples_per_second": 489.145,
      "eval_steps_per_second": 7.696,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 177.060302734375,
      "learning_rate": 9e-05,
      "loss": 0.1149,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.34156107902526855,
      "eval_runtime": 7.6548,
      "eval_samples_per_second": 489.89,
      "eval_steps_per_second": 7.708,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 205.2045135498047,
      "learning_rate": 9e-05,
      "loss": 0.0863,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9074666666666666,
      "eval_loss": 0.3585382103919983,
      "eval_runtime": 7.6697,
      "eval_samples_per_second": 488.937,
      "eval_steps_per_second": 7.693,
      "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}