trainer_state.json

{
  "best_metric": 0.8293333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0177/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": 88.5759048461914,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 3.7274,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.41573333333333334,
      "eval_loss": 3.3749473094940186,
      "eval_runtime": 13.3447,
      "eval_samples_per_second": 281.01,
      "eval_steps_per_second": 4.421,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 110.4254150390625,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.9707,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.696,
      "eval_loss": 1.141223430633545,
      "eval_runtime": 13.4032,
      "eval_samples_per_second": 279.783,
      "eval_steps_per_second": 4.402,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 387.032958984375,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.9042,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7685333333333333,
      "eval_loss": 0.7738677263259888,
      "eval_runtime": 13.8124,
      "eval_samples_per_second": 271.496,
      "eval_steps_per_second": 4.272,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 188.49327087402344,
      "learning_rate": 2.5e-05,
      "loss": 0.678,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8098666666666666,
      "eval_loss": 0.6435986161231995,
      "eval_runtime": 13.6019,
      "eval_samples_per_second": 275.696,
      "eval_steps_per_second": 4.338,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 393.73675537109375,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.5721,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8242666666666667,
      "eval_loss": 0.58692866563797,
      "eval_runtime": 13.7196,
      "eval_samples_per_second": 273.332,
      "eval_steps_per_second": 4.3,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 411.7083435058594,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.5155,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8293333333333334,
      "eval_loss": 0.562585175037384,
      "eval_runtime": 13.2407,
      "eval_samples_per_second": 283.217,
      "eval_steps_per_second": 4.456,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}