trainer_state.json

{
  "best_metric": 0.8957333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0598/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 133.55313110351562,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.8139,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7490666666666667,
      "eval_loss": 0.9948502779006958,
      "eval_runtime": 14.5696,
      "eval_samples_per_second": 257.384,
      "eval_steps_per_second": 4.05,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 190.35076904296875,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.7249,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8405333333333334,
      "eval_loss": 0.5412801504135132,
      "eval_runtime": 14.6681,
      "eval_samples_per_second": 255.656,
      "eval_steps_per_second": 4.022,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 95.31004333496094,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.4651,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.4371265172958374,
      "eval_runtime": 14.5392,
      "eval_samples_per_second": 257.923,
      "eval_steps_per_second": 4.058,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 175.82980346679688,
      "learning_rate": 5e-05,
      "loss": 0.3599,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.3904901146888733,
      "eval_runtime": 14.4995,
      "eval_samples_per_second": 258.629,
      "eval_steps_per_second": 4.069,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 107.29396057128906,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.2925,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.888,
      "eval_loss": 0.374313086271286,
      "eval_runtime": 14.4908,
      "eval_samples_per_second": 258.785,
      "eval_steps_per_second": 4.072,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 52.07404327392578,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.2493,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.35402819514274597,
      "eval_runtime": 14.4887,
      "eval_samples_per_second": 258.823,
      "eval_steps_per_second": 4.072,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 158.13624572753906,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.2318,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8957333333333334,
      "eval_loss": 0.35108017921447754,
      "eval_runtime": 14.55,
      "eval_samples_per_second": 257.732,
      "eval_steps_per_second": 4.055,
      "step": 2331
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}