trainer_state.json

{
  "best_metric": 0.9176,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0153/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": 29.365306854248047,
      "learning_rate": 0.0004,
      "loss": 1.2982,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8394666666666667,
      "eval_loss": 0.544887363910675,
      "eval_runtime": 7.7009,
      "eval_samples_per_second": 486.954,
      "eval_steps_per_second": 7.661,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 16.0965518951416,
      "learning_rate": 0.0003,
      "loss": 0.377,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8994666666666666,
      "eval_loss": 0.3504638075828552,
      "eval_runtime": 7.6707,
      "eval_samples_per_second": 488.872,
      "eval_steps_per_second": 7.692,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 29.985897064208984,
      "learning_rate": 0.0002,
      "loss": 0.2157,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9136,
      "eval_loss": 0.3115847706794739,
      "eval_runtime": 7.6819,
      "eval_samples_per_second": 488.162,
      "eval_steps_per_second": 7.68,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 43.572879791259766,
      "learning_rate": 0.0001,
      "loss": 0.1201,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.912,
      "eval_loss": 0.3125614523887634,
      "eval_runtime": 7.7163,
      "eval_samples_per_second": 485.985,
      "eval_steps_per_second": 7.646,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 38.380126953125,
      "learning_rate": 0.0,
      "loss": 0.0769,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9176,
      "eval_loss": 0.30027785897254944,
      "eval_runtime": 7.741,
      "eval_samples_per_second": 484.436,
      "eval_steps_per_second": 7.622,
      "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
}