trainer_state.json

{
  "best_metric": 0.8792,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0453/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 238.29595947265625,
      "learning_rate": 5e-05,
      "loss": 3.6907,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4978666666666667,
      "eval_loss": 3.0891404151916504,
      "eval_runtime": 15.1586,
      "eval_samples_per_second": 247.384,
      "eval_steps_per_second": 3.892,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 192.36843872070312,
      "learning_rate": 5e-05,
      "loss": 1.6748,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7597333333333334,
      "eval_loss": 0.9038434624671936,
      "eval_runtime": 15.2176,
      "eval_samples_per_second": 246.425,
      "eval_steps_per_second": 3.877,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 189.37110900878906,
      "learning_rate": 5e-05,
      "loss": 0.7571,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8186666666666667,
      "eval_loss": 0.6326377391815186,
      "eval_runtime": 15.1416,
      "eval_samples_per_second": 247.661,
      "eval_steps_per_second": 3.897,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 187.07815551757812,
      "learning_rate": 5e-05,
      "loss": 0.556,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8514666666666667,
      "eval_loss": 0.4996899664402008,
      "eval_runtime": 15.1089,
      "eval_samples_per_second": 248.198,
      "eval_steps_per_second": 3.905,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 93.12608337402344,
      "learning_rate": 5e-05,
      "loss": 0.4301,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8621333333333333,
      "eval_loss": 0.47072672843933105,
      "eval_runtime": 16.1245,
      "eval_samples_per_second": 232.565,
      "eval_steps_per_second": 3.659,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 96.14740753173828,
      "learning_rate": 5e-05,
      "loss": 0.351,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.4147716760635376,
      "eval_runtime": 14.9419,
      "eval_samples_per_second": 250.972,
      "eval_steps_per_second": 3.949,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}