trainer_state.json

{
  "best_metric": 0.908,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0706/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": 100.32328796386719,
      "learning_rate": 0.0001,
      "loss": 2.5653,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7994666666666667,
      "eval_loss": 0.7579194903373718,
      "eval_runtime": 8.0175,
      "eval_samples_per_second": 467.728,
      "eval_steps_per_second": 7.359,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 65.50557708740234,
      "learning_rate": 0.0001,
      "loss": 0.5543,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.4330437481403351,
      "eval_runtime": 7.8042,
      "eval_samples_per_second": 480.509,
      "eval_steps_per_second": 7.56,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 73.88442993164062,
      "learning_rate": 0.0001,
      "loss": 0.3332,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.3763604760169983,
      "eval_runtime": 7.4056,
      "eval_samples_per_second": 506.372,
      "eval_steps_per_second": 7.967,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 55.44818878173828,
      "learning_rate": 0.0001,
      "loss": 0.2198,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.34288695454597473,
      "eval_runtime": 7.428,
      "eval_samples_per_second": 504.849,
      "eval_steps_per_second": 7.943,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 80.1574935913086,
      "learning_rate": 0.0001,
      "loss": 0.1643,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.908,
      "eval_loss": 0.3257678747177124,
      "eval_runtime": 7.3824,
      "eval_samples_per_second": 507.967,
      "eval_steps_per_second": 7.992,
      "step": 1665
    }
  ],
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}