trainer_state.json

{
  "best_metric": 0.8653333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0433/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": 51.13264083862305,
      "learning_rate": 0.0001,
      "loss": 2.8773,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6712,
      "eval_loss": 1.1681334972381592,
      "eval_runtime": 15.2455,
      "eval_samples_per_second": 245.975,
      "eval_steps_per_second": 3.87,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 78.63920593261719,
      "learning_rate": 0.0001,
      "loss": 0.8553,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8026666666666666,
      "eval_loss": 0.6475529670715332,
      "eval_runtime": 15.1461,
      "eval_samples_per_second": 247.589,
      "eval_steps_per_second": 3.895,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 71.2088623046875,
      "learning_rate": 0.0001,
      "loss": 0.5485,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8282666666666667,
      "eval_loss": 0.5413275957107544,
      "eval_runtime": 15.0594,
      "eval_samples_per_second": 249.015,
      "eval_steps_per_second": 3.918,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 95.44227600097656,
      "learning_rate": 0.0001,
      "loss": 0.405,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8450666666666666,
      "eval_loss": 0.47777116298675537,
      "eval_runtime": 15.024,
      "eval_samples_per_second": 249.6,
      "eval_steps_per_second": 3.927,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 37.00394058227539,
      "learning_rate": 0.0001,
      "loss": 0.3015,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.4733203649520874,
      "eval_runtime": 14.9998,
      "eval_samples_per_second": 250.003,
      "eval_steps_per_second": 3.933,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 101.6033935546875,
      "learning_rate": 0.0001,
      "loss": 0.2341,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.4421366751194,
      "eval_runtime": 15.1407,
      "eval_samples_per_second": 247.677,
      "eval_steps_per_second": 3.897,
      "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
}