trainer_state.json

{
  "best_metric": 0.9029333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0008/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": 59.86603927612305,
      "learning_rate": 0.000288581929876693,
      "loss": 1.624,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8237333333333333,
      "eval_loss": 0.583684504032135,
      "eval_runtime": 7.832,
      "eval_samples_per_second": 478.805,
      "eval_steps_per_second": 7.533,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 63.662654876708984,
      "learning_rate": 0.00025606601717798207,
      "loss": 0.4742,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8565333333333334,
      "eval_loss": 0.4867258667945862,
      "eval_runtime": 8.2164,
      "eval_samples_per_second": 456.402,
      "eval_steps_per_second": 7.181,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 42.68619918823242,
      "learning_rate": 0.00020740251485476345,
      "loss": 0.28,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.41443905234336853,
      "eval_runtime": 7.6926,
      "eval_samples_per_second": 487.484,
      "eval_steps_per_second": 7.67,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 50.763031005859375,
      "learning_rate": 0.00015,
      "loss": 0.1622,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.4211158752441406,
      "eval_runtime": 7.725,
      "eval_samples_per_second": 485.438,
      "eval_steps_per_second": 7.638,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 53.545230865478516,
      "learning_rate": 9.259748514523653e-05,
      "loss": 0.1069,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8986666666666666,
      "eval_loss": 0.37167981266975403,
      "eval_runtime": 7.8889,
      "eval_samples_per_second": 475.349,
      "eval_steps_per_second": 7.479,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 50.8946418762207,
      "learning_rate": 4.3933982822017876e-05,
      "loss": 0.074,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8984,
      "eval_loss": 0.364747554063797,
      "eval_runtime": 7.8642,
      "eval_samples_per_second": 476.843,
      "eval_steps_per_second": 7.502,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 28.875621795654297,
      "learning_rate": 1.1418070123306989e-05,
      "loss": 0.0571,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8997333333333334,
      "eval_loss": 0.3732096552848816,
      "eval_runtime": 7.4534,
      "eval_samples_per_second": 503.126,
      "eval_steps_per_second": 7.916,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 42.82817459106445,
      "learning_rate": 0.0,
      "loss": 0.0452,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9029333333333334,
      "eval_loss": 0.35775065422058105,
      "eval_runtime": 7.878,
      "eval_samples_per_second": 476.009,
      "eval_steps_per_second": 7.489,
      "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
}