trainer_state.json

{
  "best_metric": 0.8381333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0377/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": 84.33055877685547,
      "learning_rate": 4.522542485937369e-05,
      "loss": 3.5972,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5978666666666667,
      "eval_loss": 2.7622671127319336,
      "eval_runtime": 7.9212,
      "eval_samples_per_second": 473.414,
      "eval_steps_per_second": 7.448,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 189.26405334472656,
      "learning_rate": 3.272542485937369e-05,
      "loss": 1.5256,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7842666666666667,
      "eval_loss": 0.8658644556999207,
      "eval_runtime": 7.0853,
      "eval_samples_per_second": 529.268,
      "eval_steps_per_second": 8.327,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 126.33458709716797,
      "learning_rate": 1.7274575140626318e-05,
      "loss": 0.7151,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.828,
      "eval_loss": 0.6357949376106262,
      "eval_runtime": 7.1788,
      "eval_samples_per_second": 522.372,
      "eval_steps_per_second": 8.219,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 49.47602844238281,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.5735,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.5952176451683044,
      "eval_runtime": 7.0902,
      "eval_samples_per_second": 528.902,
      "eval_steps_per_second": 8.321,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 513.78759765625,
      "learning_rate": 0.0,
      "loss": 0.5314,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8381333333333333,
      "eval_loss": 0.5793869495391846,
      "eval_runtime": 7.0649,
      "eval_samples_per_second": 530.796,
      "eval_steps_per_second": 8.351,
      "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
}