trainer_state.json

{
  "best_metric": 0.8957333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0688/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": 198.51666259765625,
      "learning_rate": 5e-05,
      "loss": 3.6637,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5504,
      "eval_loss": 2.926785945892334,
      "eval_runtime": 5.2182,
      "eval_samples_per_second": 718.633,
      "eval_steps_per_second": 11.306,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 111.41118621826172,
      "learning_rate": 5e-05,
      "loss": 1.6574,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7901333333333334,
      "eval_loss": 0.8478783965110779,
      "eval_runtime": 4.8998,
      "eval_samples_per_second": 765.337,
      "eval_steps_per_second": 12.041,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 88.56995391845703,
      "learning_rate": 5e-05,
      "loss": 0.6917,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.5510861873626709,
      "eval_runtime": 4.9398,
      "eval_samples_per_second": 759.144,
      "eval_steps_per_second": 11.944,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 177.9619140625,
      "learning_rate": 5e-05,
      "loss": 0.451,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4502100348472595,
      "eval_runtime": 4.9749,
      "eval_samples_per_second": 753.78,
      "eval_steps_per_second": 11.859,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 214.54348754882812,
      "learning_rate": 5e-05,
      "loss": 0.3255,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8824,
      "eval_loss": 0.3990219533443451,
      "eval_runtime": 5.0202,
      "eval_samples_per_second": 746.987,
      "eval_steps_per_second": 11.753,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 116.00785827636719,
      "learning_rate": 5e-05,
      "loss": 0.239,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8898666666666667,
      "eval_loss": 0.38604414463043213,
      "eval_runtime": 5.0814,
      "eval_samples_per_second": 737.981,
      "eval_steps_per_second": 11.611,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 95.78466033935547,
      "learning_rate": 5e-05,
      "loss": 0.1755,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8957333333333334,
      "eval_loss": 0.36163416504859924,
      "eval_runtime": 4.9498,
      "eval_samples_per_second": 757.604,
      "eval_steps_per_second": 11.92,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}