trainer_state.json

{
  "best_metric": 0.8776,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0076/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": 320.2349853515625,
      "learning_rate": 5e-05,
      "loss": 3.7114,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4301333333333333,
      "eval_loss": 3.183850049972534,
      "eval_runtime": 7.5199,
      "eval_samples_per_second": 498.678,
      "eval_steps_per_second": 7.846,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 154.48365783691406,
      "learning_rate": 5e-05,
      "loss": 1.8168,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7453333333333333,
      "eval_loss": 0.9348368644714355,
      "eval_runtime": 7.5308,
      "eval_samples_per_second": 497.955,
      "eval_steps_per_second": 7.834,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 76.5516357421875,
      "learning_rate": 5e-05,
      "loss": 0.8233,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8232,
      "eval_loss": 0.612225353717804,
      "eval_runtime": 7.4728,
      "eval_samples_per_second": 501.823,
      "eval_steps_per_second": 7.895,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 52.131019592285156,
      "learning_rate": 5e-05,
      "loss": 0.5952,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5211203694343567,
      "eval_runtime": 7.4987,
      "eval_samples_per_second": 500.089,
      "eval_steps_per_second": 7.868,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 74.33584594726562,
      "learning_rate": 5e-05,
      "loss": 0.4757,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.4445667862892151,
      "eval_runtime": 7.5441,
      "eval_samples_per_second": 497.08,
      "eval_steps_per_second": 7.821,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 79.33649444580078,
      "learning_rate": 5e-05,
      "loss": 0.4,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8650666666666667,
      "eval_loss": 0.4358771741390228,
      "eval_runtime": 7.5433,
      "eval_samples_per_second": 497.129,
      "eval_steps_per_second": 7.822,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 50.609947204589844,
      "learning_rate": 5e-05,
      "loss": 0.3432,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.4023030698299408,
      "eval_runtime": 7.5101,
      "eval_samples_per_second": 499.329,
      "eval_steps_per_second": 7.856,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 89.07440185546875,
      "learning_rate": 5e-05,
      "loss": 0.2985,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.4007119834423065,
      "eval_runtime": 7.526,
      "eval_samples_per_second": 498.274,
      "eval_steps_per_second": 7.84,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}