trainer_state.json

{
  "best_metric": 0.8829333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0538/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 33.34249496459961,
      "learning_rate": 0.0005,
      "loss": 1.2749,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8136,
      "eval_loss": 0.593638002872467,
      "eval_runtime": 8.8737,
      "eval_samples_per_second": 422.599,
      "eval_steps_per_second": 6.649,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.67953109741211,
      "learning_rate": 0.0005,
      "loss": 0.4554,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.439903199672699,
      "eval_runtime": 8.7782,
      "eval_samples_per_second": 427.194,
      "eval_steps_per_second": 6.721,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 18.25173568725586,
      "learning_rate": 0.0005,
      "loss": 0.3034,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.45186159014701843,
      "eval_runtime": 8.7409,
      "eval_samples_per_second": 429.018,
      "eval_steps_per_second": 6.75,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 26.695377349853516,
      "learning_rate": 0.0005,
      "loss": 0.2195,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.47704780101776123,
      "eval_runtime": 8.8554,
      "eval_samples_per_second": 423.473,
      "eval_steps_per_second": 6.663,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 16.018360137939453,
      "learning_rate": 0.0005,
      "loss": 0.1662,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4901309907436371,
      "eval_runtime": 8.793,
      "eval_samples_per_second": 426.473,
      "eval_steps_per_second": 6.71,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 13.674880981445312,
      "learning_rate": 0.0005,
      "loss": 0.1314,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.4527912735939026,
      "eval_runtime": 8.807,
      "eval_samples_per_second": 425.796,
      "eval_steps_per_second": 6.699,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 16.61445426940918,
      "learning_rate": 0.0005,
      "loss": 0.0999,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8829333333333333,
      "eval_loss": 0.47210872173309326,
      "eval_runtime": 8.8075,
      "eval_samples_per_second": 425.773,
      "eval_steps_per_second": 6.699,
      "step": 2331
    }
  ],
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}