trainer_state.json

{
  "best_metric": 0.8821333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0623/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": 133.32470703125,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.9117,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7133333333333334,
      "eval_loss": 1.1515904664993286,
      "eval_runtime": 8.7113,
      "eval_samples_per_second": 430.477,
      "eval_steps_per_second": 6.773,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 89.98456573486328,
      "learning_rate": 6e-05,
      "loss": 0.7558,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8357333333333333,
      "eval_loss": 0.5712976455688477,
      "eval_runtime": 7.248,
      "eval_samples_per_second": 517.385,
      "eval_steps_per_second": 8.14,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 105.52616882324219,
      "learning_rate": 4.5e-05,
      "loss": 0.4599,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.86,
      "eval_loss": 0.45348772406578064,
      "eval_runtime": 7.2036,
      "eval_samples_per_second": 520.572,
      "eval_steps_per_second": 8.19,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 74.37884521484375,
      "learning_rate": 3e-05,
      "loss": 0.343,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8685333333333334,
      "eval_loss": 0.43210917711257935,
      "eval_runtime": 7.1504,
      "eval_samples_per_second": 524.447,
      "eval_steps_per_second": 8.251,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 60.071434020996094,
      "learning_rate": 1.5e-05,
      "loss": 0.2851,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.40274766087532043,
      "eval_runtime": 7.274,
      "eval_samples_per_second": 515.534,
      "eval_steps_per_second": 8.111,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 131.18663024902344,
      "learning_rate": 0.0,
      "loss": 0.254,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.37991514801979065,
      "eval_runtime": 7.1407,
      "eval_samples_per_second": 525.162,
      "eval_steps_per_second": 8.263,
      "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
}