trainer_state.json

{
  "best_metric": 0.9056,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0128/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": 36.791351318359375,
      "learning_rate": 0.0004375,
      "loss": 1.388,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8304,
      "eval_loss": 0.5507839918136597,
      "eval_runtime": 15.3534,
      "eval_samples_per_second": 244.245,
      "eval_steps_per_second": 3.843,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 42.965126037597656,
      "learning_rate": 0.000375,
      "loss": 0.4652,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8634666666666667,
      "eval_loss": 0.4628974497318268,
      "eval_runtime": 15.2518,
      "eval_samples_per_second": 245.873,
      "eval_steps_per_second": 3.868,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 26.47138023376465,
      "learning_rate": 0.0003125,
      "loss": 0.2838,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.44222792983055115,
      "eval_runtime": 15.3067,
      "eval_samples_per_second": 244.991,
      "eval_steps_per_second": 3.855,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 28.633764266967773,
      "learning_rate": 0.00025,
      "loss": 0.178,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.4824122488498688,
      "eval_runtime": 15.2613,
      "eval_samples_per_second": 245.72,
      "eval_steps_per_second": 3.866,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 29.588836669921875,
      "learning_rate": 0.0001875,
      "loss": 0.122,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.41775253415107727,
      "eval_runtime": 15.3149,
      "eval_samples_per_second": 244.859,
      "eval_steps_per_second": 3.852,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 38.82207489013672,
      "learning_rate": 0.000125,
      "loss": 0.066,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8954666666666666,
      "eval_loss": 0.4248058795928955,
      "eval_runtime": 15.2662,
      "eval_samples_per_second": 245.641,
      "eval_steps_per_second": 3.865,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 40.97157287597656,
      "learning_rate": 6.25e-05,
      "loss": 0.0458,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8938666666666667,
      "eval_loss": 0.4332111179828644,
      "eval_runtime": 15.262,
      "eval_samples_per_second": 245.709,
      "eval_steps_per_second": 3.866,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 31.818134307861328,
      "learning_rate": 0.0,
      "loss": 0.0293,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9056,
      "eval_loss": 0.4111049771308899,
      "eval_runtime": 15.2254,
      "eval_samples_per_second": 246.299,
      "eval_steps_per_second": 3.875,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}