trainer_state.json

{
  "best_metric": 0.808,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0317/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 70.31110382080078,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 3.8236,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.41013333333333335,
      "eval_loss": 3.6402487754821777,
      "eval_runtime": 7.9247,
      "eval_samples_per_second": 473.203,
      "eval_steps_per_second": 7.445,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 116.08302307128906,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 3.1374,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5701333333333334,
      "eval_loss": 2.468851089477539,
      "eval_runtime": 7.6274,
      "eval_samples_per_second": 491.648,
      "eval_steps_per_second": 7.735,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 263.5459899902344,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 1.7379,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7117333333333333,
      "eval_loss": 1.2245771884918213,
      "eval_runtime": 7.8224,
      "eval_samples_per_second": 479.393,
      "eval_steps_per_second": 7.542,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 137.12254333496094,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.034,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7576,
      "eval_loss": 0.9176701903343201,
      "eval_runtime": 7.9682,
      "eval_samples_per_second": 470.623,
      "eval_steps_per_second": 7.404,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 147.89862060546875,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.7955,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7850666666666667,
      "eval_loss": 0.7701371908187866,
      "eval_runtime": 8.1723,
      "eval_samples_per_second": 458.87,
      "eval_steps_per_second": 7.22,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 175.3457489013672,
      "learning_rate": 9.999999999999999e-06,
      "loss": 0.6711,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7984,
      "eval_loss": 0.6954808831214905,
      "eval_runtime": 8.005,
      "eval_samples_per_second": 468.457,
      "eval_steps_per_second": 7.37,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 115.50255584716797,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.603,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8045333333333333,
      "eval_loss": 0.6646720170974731,
      "eval_runtime": 7.9429,
      "eval_samples_per_second": 472.118,
      "eval_steps_per_second": 7.428,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 308.2919921875,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.5565,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.808,
      "eval_loss": 0.660474956035614,
      "eval_runtime": 7.7579,
      "eval_samples_per_second": 483.376,
      "eval_steps_per_second": 7.605,
      "step": 2664
    }
  ],
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}