trainer_state.json

{
  "best_metric": 0.8378666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0663/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": 129.93133544921875,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 3.8301,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4437333333333333,
      "eval_loss": 3.631554365158081,
      "eval_runtime": 17.355,
      "eval_samples_per_second": 216.076,
      "eval_steps_per_second": 3.4,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 102.8958969116211,
      "learning_rate": 2.649066664678467e-05,
      "loss": 3.0036,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6594666666666666,
      "eval_loss": 2.108182191848755,
      "eval_runtime": 17.2489,
      "eval_samples_per_second": 217.405,
      "eval_steps_per_second": 3.42,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 219.96543884277344,
      "learning_rate": 2.25e-05,
      "loss": 1.4168,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7728,
      "eval_loss": 0.975724458694458,
      "eval_runtime": 16.8515,
      "eval_samples_per_second": 222.533,
      "eval_steps_per_second": 3.501,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 124.4221420288086,
      "learning_rate": 1.760472266500396e-05,
      "loss": 0.8269,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8042666666666667,
      "eval_loss": 0.750277578830719,
      "eval_runtime": 16.9461,
      "eval_samples_per_second": 221.289,
      "eval_steps_per_second": 3.482,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 213.7254638671875,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 0.6441,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8245333333333333,
      "eval_loss": 0.656435489654541,
      "eval_runtime": 16.4745,
      "eval_samples_per_second": 227.625,
      "eval_steps_per_second": 3.581,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 353.54083251953125,
      "learning_rate": 7.500000000000004e-06,
      "loss": 0.5567,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8274666666666667,
      "eval_loss": 0.6141520142555237,
      "eval_runtime": 16.2928,
      "eval_samples_per_second": 230.163,
      "eval_steps_per_second": 3.621,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 390.2662658691406,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 0.5173,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8349333333333333,
      "eval_loss": 0.5816302299499512,
      "eval_runtime": 15.0804,
      "eval_samples_per_second": 248.667,
      "eval_steps_per_second": 3.912,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 181.34442138671875,
      "learning_rate": 9.046106882113753e-07,
      "loss": 0.4887,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8378666666666666,
      "eval_loss": 0.5704259872436523,
      "eval_runtime": 14.0568,
      "eval_samples_per_second": 266.775,
      "eval_steps_per_second": 4.197,
      "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
}