trainer_state.json

{
  "best_metric": 0.8976,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0267/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": 25.308975219726562,
      "learning_rate": 0.0004444444444444444,
      "loss": 1.3709,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8149333333333333,
      "eval_loss": 0.6261569261550903,
      "eval_runtime": 8.9239,
      "eval_samples_per_second": 420.219,
      "eval_steps_per_second": 6.611,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.4732608795166,
      "learning_rate": 0.0003888888888888889,
      "loss": 0.4457,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.46363815665245056,
      "eval_runtime": 8.7689,
      "eval_samples_per_second": 427.65,
      "eval_steps_per_second": 6.728,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 34.33594512939453,
      "learning_rate": 0.0003333333333333333,
      "loss": 0.2672,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.45999643206596375,
      "eval_runtime": 8.799,
      "eval_samples_per_second": 426.184,
      "eval_steps_per_second": 6.705,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 20.232431411743164,
      "learning_rate": 0.0002777777777777778,
      "loss": 0.1555,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.4661049544811249,
      "eval_runtime": 8.7375,
      "eval_samples_per_second": 429.184,
      "eval_steps_per_second": 6.752,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 22.158390045166016,
      "learning_rate": 0.0002222222222222222,
      "loss": 0.0998,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8789333333333333,
      "eval_loss": 0.4818182587623596,
      "eval_runtime": 8.5756,
      "eval_samples_per_second": 437.288,
      "eval_steps_per_second": 6.88,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 59.91155242919922,
      "learning_rate": 0.00016666666666666666,
      "loss": 0.0646,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.5086092948913574,
      "eval_runtime": 8.7681,
      "eval_samples_per_second": 427.688,
      "eval_steps_per_second": 6.729,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 31.50436782836914,
      "learning_rate": 0.0001111111111111111,
      "loss": 0.0464,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.47028401494026184,
      "eval_runtime": 8.5362,
      "eval_samples_per_second": 439.307,
      "eval_steps_per_second": 6.912,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 23.63042449951172,
      "learning_rate": 5.555555555555555e-05,
      "loss": 0.0226,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8976,
      "eval_loss": 0.4578145742416382,
      "eval_runtime": 8.691,
      "eval_samples_per_second": 431.483,
      "eval_steps_per_second": 6.789,
      "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
}