trainer_state.json

{
  "best_metric": 0.904,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0696/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": 222.6498260498047,
      "learning_rate": 0.0001,
      "loss": 2.7416,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7736,
      "eval_loss": 0.8790609240531921,
      "eval_runtime": 7.9594,
      "eval_samples_per_second": 471.141,
      "eval_steps_per_second": 7.413,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 137.5121307373047,
      "learning_rate": 0.0001,
      "loss": 0.689,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.4691271483898163,
      "eval_runtime": 7.8775,
      "eval_samples_per_second": 476.042,
      "eval_steps_per_second": 7.49,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 122.9374771118164,
      "learning_rate": 0.0001,
      "loss": 0.444,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4077308475971222,
      "eval_runtime": 7.5929,
      "eval_samples_per_second": 493.881,
      "eval_steps_per_second": 7.77,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 62.6981201171875,
      "learning_rate": 0.0001,
      "loss": 0.3326,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.36339330673217773,
      "eval_runtime": 7.7016,
      "eval_samples_per_second": 486.91,
      "eval_steps_per_second": 7.661,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 148.9110870361328,
      "learning_rate": 0.0001,
      "loss": 0.2596,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.904,
      "eval_loss": 0.32136058807373047,
      "eval_runtime": 8.0167,
      "eval_samples_per_second": 467.776,
      "eval_steps_per_second": 7.36,
      "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
}