trainer_state.json

{
  "best_metric": 0.9045333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0620/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": 46.615657806396484,
      "learning_rate": 0.0003,
      "loss": 1.4823,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8450666666666666,
      "eval_loss": 0.5198315978050232,
      "eval_runtime": 12.7234,
      "eval_samples_per_second": 294.733,
      "eval_steps_per_second": 4.637,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 42.4277229309082,
      "learning_rate": 0.0003,
      "loss": 0.4129,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.4055810570716858,
      "eval_runtime": 12.7819,
      "eval_samples_per_second": 293.383,
      "eval_steps_per_second": 4.616,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 34.12581253051758,
      "learning_rate": 0.0003,
      "loss": 0.25,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.36583125591278076,
      "eval_runtime": 12.8046,
      "eval_samples_per_second": 292.862,
      "eval_steps_per_second": 4.608,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 34.073020935058594,
      "learning_rate": 0.0003,
      "loss": 0.1579,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.3578318953514099,
      "eval_runtime": 12.8573,
      "eval_samples_per_second": 291.663,
      "eval_steps_per_second": 4.589,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 24.514530181884766,
      "learning_rate": 0.0003,
      "loss": 0.1221,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.3876437246799469,
      "eval_runtime": 12.7118,
      "eval_samples_per_second": 295.001,
      "eval_steps_per_second": 4.641,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 16.45797348022461,
      "learning_rate": 0.0003,
      "loss": 0.0874,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8976,
      "eval_loss": 0.4097873568534851,
      "eval_runtime": 12.7123,
      "eval_samples_per_second": 294.989,
      "eval_steps_per_second": 4.641,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 45.04862594604492,
      "learning_rate": 0.0003,
      "loss": 0.0744,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8978666666666667,
      "eval_loss": 0.45026713609695435,
      "eval_runtime": 12.9266,
      "eval_samples_per_second": 290.1,
      "eval_steps_per_second": 4.564,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 30.30103874206543,
      "learning_rate": 0.0003,
      "loss": 0.0619,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9045333333333333,
      "eval_loss": 0.4282241463661194,
      "eval_runtime": 12.9093,
      "eval_samples_per_second": 290.487,
      "eval_steps_per_second": 4.57,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}