trainer_state.json

{
  "best_metric": 0.8352,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0441/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": 180.93121337890625,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 3.6325,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5712,
      "eval_loss": 2.973581075668335,
      "eval_runtime": 7.6642,
      "eval_samples_per_second": 489.289,
      "eval_steps_per_second": 7.698,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 122.09307861328125,
      "learning_rate": 3.571428571428572e-05,
      "loss": 1.8694,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7368,
      "eval_loss": 1.044941782951355,
      "eval_runtime": 7.6033,
      "eval_samples_per_second": 493.209,
      "eval_steps_per_second": 7.76,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 289.08477783203125,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.8445,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.7115259170532227,
      "eval_runtime": 7.6401,
      "eval_samples_per_second": 490.829,
      "eval_steps_per_second": 7.722,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 185.44500732421875,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 0.6205,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8272,
      "eval_loss": 0.5869815349578857,
      "eval_runtime": 7.5812,
      "eval_samples_per_second": 494.644,
      "eval_steps_per_second": 7.782,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 183.4117431640625,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.5177,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8333333333333334,
      "eval_loss": 0.5515949130058289,
      "eval_runtime": 7.6112,
      "eval_samples_per_second": 492.694,
      "eval_steps_per_second": 7.752,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 117.58324432373047,
      "learning_rate": 7.142857142857143e-06,
      "loss": 0.4573,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8352,
      "eval_loss": 0.5361629724502563,
      "eval_runtime": 7.6646,
      "eval_samples_per_second": 489.26,
      "eval_steps_per_second": 7.698,
      "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
}