trainer_state.json

{
  "best_metric": 0.8890666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0205/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": 130.36505126953125,
      "learning_rate": 6.788924172750679e-05,
      "loss": 3.2281,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6717333333333333,
      "eval_loss": 1.5886704921722412,
      "eval_runtime": 8.3878,
      "eval_samples_per_second": 447.08,
      "eval_steps_per_second": 7.034,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 95.63096618652344,
      "learning_rate": 6.181155550916422e-05,
      "loss": 0.8706,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8253333333333334,
      "eval_loss": 0.5777832269668579,
      "eval_runtime": 8.3773,
      "eval_samples_per_second": 447.64,
      "eval_steps_per_second": 7.043,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 112.73579406738281,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.4982,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.4370090365409851,
      "eval_runtime": 8.1346,
      "eval_samples_per_second": 460.991,
      "eval_steps_per_second": 7.253,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 154.8848419189453,
      "learning_rate": 4.1077686218342566e-05,
      "loss": 0.3895,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8733333333333333,
      "eval_loss": 0.38617685437202454,
      "eval_runtime": 8.2463,
      "eval_samples_per_second": 454.751,
      "eval_steps_per_second": 7.155,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 83.54861450195312,
      "learning_rate": 2.8922313781657438e-05,
      "loss": 0.3144,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.3751835227012634,
      "eval_runtime": 8.3459,
      "eval_samples_per_second": 449.321,
      "eval_steps_per_second": 7.069,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 78.3084716796875,
      "learning_rate": 1.7500000000000005e-05,
      "loss": 0.2653,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8832,
      "eval_loss": 0.36916041374206543,
      "eval_runtime": 8.2898,
      "eval_samples_per_second": 452.363,
      "eval_steps_per_second": 7.117,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 110.19779968261719,
      "learning_rate": 8.188444490835772e-06,
      "loss": 0.2376,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.888,
      "eval_loss": 0.35475069284439087,
      "eval_runtime": 7.9784,
      "eval_samples_per_second": 470.018,
      "eval_steps_per_second": 7.395,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 252.7132568359375,
      "learning_rate": 2.1107582724932087e-06,
      "loss": 0.231,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8890666666666667,
      "eval_loss": 0.3485003113746643,
      "eval_runtime": 8.2601,
      "eval_samples_per_second": 453.989,
      "eval_steps_per_second": 7.143,
      "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
}