trainer_state.json

{
  "best_metric": 0.8978666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0053/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": 101.83961486816406,
      "learning_rate": 7e-05,
      "loss": 3.2501,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6813333333333333,
      "eval_loss": 1.7394237518310547,
      "eval_runtime": 7.6625,
      "eval_samples_per_second": 489.394,
      "eval_steps_per_second": 7.7,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 171.45413208007812,
      "learning_rate": 7e-05,
      "loss": 0.878,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8434666666666667,
      "eval_loss": 0.5574154853820801,
      "eval_runtime": 7.5761,
      "eval_samples_per_second": 494.978,
      "eval_steps_per_second": 7.788,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 145.1888885498047,
      "learning_rate": 7e-05,
      "loss": 0.4275,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.4259926974773407,
      "eval_runtime": 7.6064,
      "eval_samples_per_second": 493.009,
      "eval_steps_per_second": 7.757,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 261.71771240234375,
      "learning_rate": 7e-05,
      "loss": 0.2818,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.38107460737228394,
      "eval_runtime": 7.5721,
      "eval_samples_per_second": 495.237,
      "eval_steps_per_second": 7.792,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 158.42453002929688,
      "learning_rate": 7e-05,
      "loss": 0.1983,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8944,
      "eval_loss": 0.3571372330188751,
      "eval_runtime": 7.6064,
      "eval_samples_per_second": 493.007,
      "eval_steps_per_second": 7.757,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 78.93113708496094,
      "learning_rate": 7e-05,
      "loss": 0.1392,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8978666666666667,
      "eval_loss": 0.3535504639148712,
      "eval_runtime": 7.6198,
      "eval_samples_per_second": 492.142,
      "eval_steps_per_second": 7.743,
      "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
}