trainer_state.json

{
  "best_metric": 0.8725333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0141/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": 102.29676818847656,
      "learning_rate": 5.833333333333333e-05,
      "loss": 3.2997,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.652,
      "eval_loss": 1.7398937940597534,
      "eval_runtime": 14.5084,
      "eval_samples_per_second": 258.471,
      "eval_steps_per_second": 4.067,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 198.0923309326172,
      "learning_rate": 4.666666666666666e-05,
      "loss": 0.9999,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8288,
      "eval_loss": 0.6047856211662292,
      "eval_runtime": 14.3653,
      "eval_samples_per_second": 261.045,
      "eval_steps_per_second": 4.107,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 197.35838317871094,
      "learning_rate": 3.5e-05,
      "loss": 0.5297,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.4969188868999481,
      "eval_runtime": 14.3214,
      "eval_samples_per_second": 261.845,
      "eval_steps_per_second": 4.12,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 147.95130920410156,
      "learning_rate": 2.333333333333333e-05,
      "loss": 0.3824,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8650666666666667,
      "eval_loss": 0.450914204120636,
      "eval_runtime": 14.3367,
      "eval_samples_per_second": 261.566,
      "eval_steps_per_second": 4.115,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 148.26661682128906,
      "learning_rate": 1.1666666666666665e-05,
      "loss": 0.3111,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.42940637469291687,
      "eval_runtime": 14.2848,
      "eval_samples_per_second": 262.517,
      "eval_steps_per_second": 4.13,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 119.8072280883789,
      "learning_rate": 0.0,
      "loss": 0.2695,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8725333333333334,
      "eval_loss": 0.4119669497013092,
      "eval_runtime": 14.5605,
      "eval_samples_per_second": 257.546,
      "eval_steps_per_second": 4.052,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}