trainer_state.json

{
  "best_metric": 0.8930666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0006/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": 207.81982421875,
      "learning_rate": 7e-05,
      "loss": 3.0642,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6984,
      "eval_loss": 1.2565984725952148,
      "eval_runtime": 14.1762,
      "eval_samples_per_second": 264.527,
      "eval_steps_per_second": 4.162,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 69.64102172851562,
      "learning_rate": 7e-05,
      "loss": 0.8022,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8301333333333333,
      "eval_loss": 0.5640637278556824,
      "eval_runtime": 13.7973,
      "eval_samples_per_second": 271.792,
      "eval_steps_per_second": 4.276,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 174.96966552734375,
      "learning_rate": 7e-05,
      "loss": 0.4432,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.45328691601753235,
      "eval_runtime": 13.9366,
      "eval_samples_per_second": 269.076,
      "eval_steps_per_second": 4.233,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 87.76078796386719,
      "learning_rate": 7e-05,
      "loss": 0.2971,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8810666666666667,
      "eval_loss": 0.3990769386291504,
      "eval_runtime": 13.3464,
      "eval_samples_per_second": 280.975,
      "eval_steps_per_second": 4.421,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 63.60320281982422,
      "learning_rate": 7e-05,
      "loss": 0.2039,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.4048590362071991,
      "eval_runtime": 13.4412,
      "eval_samples_per_second": 278.993,
      "eval_steps_per_second": 4.389,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 199.33486938476562,
      "learning_rate": 7e-05,
      "loss": 0.1454,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.38799089193344116,
      "eval_runtime": 13.3087,
      "eval_samples_per_second": 281.771,
      "eval_steps_per_second": 4.433,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 140.1734161376953,
      "learning_rate": 7e-05,
      "loss": 0.0994,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.42011451721191406,
      "eval_runtime": 13.2851,
      "eval_samples_per_second": 282.271,
      "eval_steps_per_second": 4.441,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 95.13492584228516,
      "learning_rate": 7e-05,
      "loss": 0.0719,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.41516998410224915,
      "eval_runtime": 13.3095,
      "eval_samples_per_second": 281.754,
      "eval_steps_per_second": 4.433,
      "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
}