trainer_state.json

{
  "best_metric": 0.852,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0100/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": 122.592041015625,
      "learning_rate": 3e-05,
      "loss": 3.8322,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.42133333333333334,
      "eval_loss": 3.687612771987915,
      "eval_runtime": 7.3051,
      "eval_samples_per_second": 513.342,
      "eval_steps_per_second": 8.077,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 193.00152587890625,
      "learning_rate": 3e-05,
      "loss": 2.9992,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6512,
      "eval_loss": 2.038257360458374,
      "eval_runtime": 7.596,
      "eval_samples_per_second": 493.678,
      "eval_steps_per_second": 7.767,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 176.11065673828125,
      "learning_rate": 3e-05,
      "loss": 1.3437,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7810666666666667,
      "eval_loss": 0.8375335931777954,
      "eval_runtime": 7.1993,
      "eval_samples_per_second": 520.885,
      "eval_steps_per_second": 8.195,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 194.34048461914062,
      "learning_rate": 3e-05,
      "loss": 0.724,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8224,
      "eval_loss": 0.600430965423584,
      "eval_runtime": 7.2082,
      "eval_samples_per_second": 520.242,
      "eval_steps_per_second": 8.185,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 318.91693115234375,
      "learning_rate": 3e-05,
      "loss": 0.5247,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8445333333333334,
      "eval_loss": 0.5087047815322876,
      "eval_runtime": 7.464,
      "eval_samples_per_second": 502.415,
      "eval_steps_per_second": 7.905,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 175.8058624267578,
      "learning_rate": 3e-05,
      "loss": 0.4351,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.45084869861602783,
      "eval_runtime": 7.1396,
      "eval_samples_per_second": 525.239,
      "eval_steps_per_second": 8.264,
      "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
}