trainer_state.json

{
  "best_metric": 0.8794666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0537/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": 78.32310485839844,
      "learning_rate": 9e-05,
      "loss": 2.9807,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7165333333333334,
      "eval_loss": 1.193550944328308,
      "eval_runtime": 15.4105,
      "eval_samples_per_second": 243.34,
      "eval_steps_per_second": 3.829,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 233.14247131347656,
      "learning_rate": 9e-05,
      "loss": 0.8332,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8277333333333333,
      "eval_loss": 0.6076503992080688,
      "eval_runtime": 15.5026,
      "eval_samples_per_second": 241.895,
      "eval_steps_per_second": 3.806,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 179.40504455566406,
      "learning_rate": 9e-05,
      "loss": 0.507,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.4708832800388336,
      "eval_runtime": 15.3195,
      "eval_samples_per_second": 244.786,
      "eval_steps_per_second": 3.851,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 121.13370513916016,
      "learning_rate": 9e-05,
      "loss": 0.3648,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.4190102815628052,
      "eval_runtime": 15.3813,
      "eval_samples_per_second": 243.803,
      "eval_steps_per_second": 3.836,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 50.28668212890625,
      "learning_rate": 9e-05,
      "loss": 0.2783,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.40447503328323364,
      "eval_runtime": 15.3746,
      "eval_samples_per_second": 243.909,
      "eval_steps_per_second": 3.838,
      "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
}