trainer_state.json

{
  "best_metric": 0.8810666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0115/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": 85.74459075927734,
      "learning_rate": 8.397114317029975e-05,
      "loss": 2.9191,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7426666666666667,
      "eval_loss": 1.0046042203903198,
      "eval_runtime": 14.6663,
      "eval_samples_per_second": 255.688,
      "eval_steps_per_second": 4.023,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 199.49851989746094,
      "learning_rate": 6.75e-05,
      "loss": 0.7542,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8424,
      "eval_loss": 0.5256293416023254,
      "eval_runtime": 14.6295,
      "eval_samples_per_second": 256.331,
      "eval_steps_per_second": 4.033,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 71.75749969482422,
      "learning_rate": 4.5e-05,
      "loss": 0.4983,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.443558931350708,
      "eval_runtime": 14.6379,
      "eval_samples_per_second": 256.184,
      "eval_steps_per_second": 4.031,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 141.67465209960938,
      "learning_rate": 2.250000000000001e-05,
      "loss": 0.3876,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.88,
      "eval_loss": 0.4073806703090668,
      "eval_runtime": 14.6096,
      "eval_samples_per_second": 256.681,
      "eval_steps_per_second": 4.038,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 114.67706298828125,
      "learning_rate": 6.028856829700258e-06,
      "loss": 0.3425,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.4056820869445801,
      "eval_runtime": 14.5861,
      "eval_samples_per_second": 257.093,
      "eval_steps_per_second": 4.045,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 101.54653930664062,
      "learning_rate": 0.0,
      "loss": 0.3252,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.3936315178871155,
      "eval_runtime": 14.5971,
      "eval_samples_per_second": 256.9,
      "eval_steps_per_second": 4.042,
      "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
}