trainer_state.json

{
  "best_metric": 0.8792,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0236/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": 117.95266723632812,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 3.5634,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6170666666666667,
      "eval_loss": 2.6925179958343506,
      "eval_runtime": 7.185,
      "eval_samples_per_second": 521.923,
      "eval_steps_per_second": 8.212,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 79.44466400146484,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.473,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8032,
      "eval_loss": 0.7636287808418274,
      "eval_runtime": 7.1553,
      "eval_samples_per_second": 524.089,
      "eval_steps_per_second": 8.246,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 63.05199432373047,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.6456,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.5356379151344299,
      "eval_runtime": 7.1662,
      "eval_samples_per_second": 523.287,
      "eval_steps_per_second": 8.233,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 327.1995849609375,
      "learning_rate": 2.5e-05,
      "loss": 0.479,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8666666666666667,
      "eval_loss": 0.45373037457466125,
      "eval_runtime": 7.1709,
      "eval_samples_per_second": 522.947,
      "eval_steps_per_second": 8.228,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 175.28271484375,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.397,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8789333333333333,
      "eval_loss": 0.41331151127815247,
      "eval_runtime": 7.122,
      "eval_samples_per_second": 526.536,
      "eval_steps_per_second": 8.284,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 226.25001525878906,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.35,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4118313491344452,
      "eval_runtime": 7.0729,
      "eval_samples_per_second": 530.19,
      "eval_steps_per_second": 8.342,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 123.58414459228516,
      "learning_rate": 1.9030116872178316e-06,
      "loss": 0.3227,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8789333333333333,
      "eval_loss": 0.39352425932884216,
      "eval_runtime": 7.1075,
      "eval_samples_per_second": 527.611,
      "eval_steps_per_second": 8.301,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 369.1617126464844,
      "learning_rate": 0.0,
      "loss": 0.3184,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.398346871137619,
      "eval_runtime": 7.1042,
      "eval_samples_per_second": 527.855,
      "eval_steps_per_second": 8.305,
      "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
}