trainer_state.json

{
  "best_metric": 0.8837333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0756/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": 33.528778076171875,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.8404,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7002666666666667,
      "eval_loss": 1.1113290786743164,
      "eval_runtime": 5.3734,
      "eval_samples_per_second": 697.882,
      "eval_steps_per_second": 10.98,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 192.3518524169922,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.7962,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8274666666666667,
      "eval_loss": 0.5882378220558167,
      "eval_runtime": 5.2394,
      "eval_samples_per_second": 715.734,
      "eval_steps_per_second": 11.261,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 176.70057678222656,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.4966,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.47668713331222534,
      "eval_runtime": 5.2487,
      "eval_samples_per_second": 714.462,
      "eval_steps_per_second": 11.241,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 77.66934967041016,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.3652,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.4447057247161865,
      "eval_runtime": 5.4015,
      "eval_samples_per_second": 694.251,
      "eval_steps_per_second": 10.923,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 108.84664154052734,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.2882,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.43080785870552063,
      "eval_runtime": 5.2426,
      "eval_samples_per_second": 715.3,
      "eval_steps_per_second": 11.254,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 76.30027770996094,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.2413,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.3895798921585083,
      "eval_runtime": 5.2495,
      "eval_samples_per_second": 714.353,
      "eval_steps_per_second": 11.239,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 68.49740600585938,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.2056,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.394468754529953,
      "eval_runtime": 5.2606,
      "eval_samples_per_second": 712.849,
      "eval_steps_per_second": 11.215,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 154.12335205078125,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.1934,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.38185036182403564,
      "eval_runtime": 5.5213,
      "eval_samples_per_second": 679.185,
      "eval_steps_per_second": 10.686,
      "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
}