trainer_state.json

{
  "best_metric": 0.8792,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0009/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 300.2476501464844,
      "learning_rate": 7e-05,
      "loss": 3.3601,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6264,
      "eval_loss": 1.8871158361434937,
      "eval_runtime": 7.8443,
      "eval_samples_per_second": 478.052,
      "eval_steps_per_second": 7.521,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 78.6750717163086,
      "learning_rate": 7e-05,
      "loss": 1.0058,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8186666666666667,
      "eval_loss": 0.6207583546638489,
      "eval_runtime": 7.7625,
      "eval_samples_per_second": 483.089,
      "eval_steps_per_second": 7.601,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 109.4736328125,
      "learning_rate": 7e-05,
      "loss": 0.5246,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.5088762640953064,
      "eval_runtime": 8.0798,
      "eval_samples_per_second": 464.12,
      "eval_steps_per_second": 7.302,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 196.1301727294922,
      "learning_rate": 7e-05,
      "loss": 0.3898,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.428166002035141,
      "eval_runtime": 7.9007,
      "eval_samples_per_second": 474.642,
      "eval_steps_per_second": 7.468,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 96.71028900146484,
      "learning_rate": 7e-05,
      "loss": 0.3057,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.39524516463279724,
      "eval_runtime": 7.6127,
      "eval_samples_per_second": 492.6,
      "eval_steps_per_second": 7.75,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}