trainer_state.json

{
  "best_metric": 0.7082666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0452/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": 264.63751220703125,
      "learning_rate": 2.4e-05,
      "loss": 3.8511,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3642666666666667,
      "eval_loss": 3.7412078380584717,
      "eval_runtime": 14.5006,
      "eval_samples_per_second": 258.61,
      "eval_steps_per_second": 4.069,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 121.53596496582031,
      "learning_rate": 1.8e-05,
      "loss": 3.4061,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5624,
      "eval_loss": 3.0640437602996826,
      "eval_runtime": 14.5324,
      "eval_samples_per_second": 258.045,
      "eval_steps_per_second": 4.06,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 296.4803771972656,
      "learning_rate": 1.2e-05,
      "loss": 2.4814,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6392,
      "eval_loss": 2.0348403453826904,
      "eval_runtime": 14.5164,
      "eval_samples_per_second": 258.328,
      "eval_steps_per_second": 4.064,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 213.36410522460938,
      "learning_rate": 6e-06,
      "loss": 1.7755,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6930666666666667,
      "eval_loss": 1.5892759561538696,
      "eval_runtime": 14.5466,
      "eval_samples_per_second": 257.793,
      "eval_steps_per_second": 4.056,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 236.36134338378906,
      "learning_rate": 0.0,
      "loss": 1.5162,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7082666666666667,
      "eval_loss": 1.6133513450622559,
      "eval_runtime": 14.4951,
      "eval_samples_per_second": 258.708,
      "eval_steps_per_second": 4.07,
      "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
}