trainer_state.json

{
  "best_metric": 0.8634666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0998/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": 79.78257751464844,
      "learning_rate": 8.140576474687264e-05,
      "loss": 2.9317,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7008,
      "eval_loss": 1.2555989027023315,
      "eval_runtime": 7.7041,
      "eval_samples_per_second": 486.754,
      "eval_steps_per_second": 7.658,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 117.36102294921875,
      "learning_rate": 5.890576474687264e-05,
      "loss": 0.8297,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8429333333333333,
      "eval_loss": 0.5707444548606873,
      "eval_runtime": 7.8279,
      "eval_samples_per_second": 479.053,
      "eval_steps_per_second": 7.537,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 140.2103271484375,
      "learning_rate": 3.1094235253127374e-05,
      "loss": 0.5195,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.856,
      "eval_loss": 0.4852633476257324,
      "eval_runtime": 7.4387,
      "eval_samples_per_second": 504.12,
      "eval_steps_per_second": 7.931,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 85.63063049316406,
      "learning_rate": 8.59423525312737e-06,
      "loss": 0.4267,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.4693203270435333,
      "eval_runtime": 7.415,
      "eval_samples_per_second": 505.734,
      "eval_steps_per_second": 7.957,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 122.87202453613281,
      "learning_rate": 0.0,
      "loss": 0.3822,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8634666666666667,
      "eval_loss": 0.4681846797466278,
      "eval_runtime": 7.8264,
      "eval_samples_per_second": 479.146,
      "eval_steps_per_second": 7.539,
      "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
}