trainer_state.json

{
  "best_metric": 0.9256,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0197/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": 224.22189331054688,
      "learning_rate": 0.0001,
      "loss": 2.6256,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8181333333333334,
      "eval_loss": 0.7544289231300354,
      "eval_runtime": 7.3177,
      "eval_samples_per_second": 512.456,
      "eval_steps_per_second": 8.063,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 78.11978912353516,
      "learning_rate": 0.0001,
      "loss": 0.538,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8957333333333334,
      "eval_loss": 0.369615375995636,
      "eval_runtime": 7.285,
      "eval_samples_per_second": 514.753,
      "eval_steps_per_second": 8.099,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 99.43171691894531,
      "learning_rate": 0.0001,
      "loss": 0.2812,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9125333333333333,
      "eval_loss": 0.31008344888687134,
      "eval_runtime": 7.515,
      "eval_samples_per_second": 499.005,
      "eval_steps_per_second": 7.851,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 111.20413208007812,
      "learning_rate": 0.0001,
      "loss": 0.1626,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9210666666666667,
      "eval_loss": 0.28139087557792664,
      "eval_runtime": 7.1325,
      "eval_samples_per_second": 525.766,
      "eval_steps_per_second": 8.272,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 81.99669647216797,
      "learning_rate": 0.0001,
      "loss": 0.1002,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9256,
      "eval_loss": 0.2987283170223236,
      "eval_runtime": 7.2295,
      "eval_samples_per_second": 518.709,
      "eval_steps_per_second": 8.161,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}