trainer_state.json

{
  "best_metric": 0.8896,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0684/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 46.50367736816406,
      "learning_rate": 0.00023999999999999998,
      "loss": 1.6828,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8032,
      "eval_loss": 0.6236593723297119,
      "eval_runtime": 9.6927,
      "eval_samples_per_second": 386.891,
      "eval_steps_per_second": 6.087,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 79.04351043701172,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.4935,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.4756107032299042,
      "eval_runtime": 9.4186,
      "eval_samples_per_second": 398.146,
      "eval_steps_per_second": 6.264,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 66.12368774414062,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.3121,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8781333333333333,
      "eval_loss": 0.4084137976169586,
      "eval_runtime": 9.6784,
      "eval_samples_per_second": 387.46,
      "eval_steps_per_second": 6.096,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 52.801639556884766,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.2065,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8853333333333333,
      "eval_loss": 0.38140034675598145,
      "eval_runtime": 9.0447,
      "eval_samples_per_second": 414.608,
      "eval_steps_per_second": 6.523,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 60.097511291503906,
      "learning_rate": 0.0,
      "loss": 0.146,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8896,
      "eval_loss": 0.3672443926334381,
      "eval_runtime": 8.9913,
      "eval_samples_per_second": 417.068,
      "eval_steps_per_second": 6.562,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}