trainer_state.json

{
  "best_metric": 0.8589333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0723/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": 53.89590835571289,
      "learning_rate": 2.9095389311788626e-05,
      "loss": 3.8144,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5122666666666666,
      "eval_loss": 3.602062463760376,
      "eval_runtime": 8.4274,
      "eval_samples_per_second": 444.977,
      "eval_steps_per_second": 7.001,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 155.98001098632812,
      "learning_rate": 2.649066664678467e-05,
      "loss": 2.9697,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6533333333333333,
      "eval_loss": 2.163749933242798,
      "eval_runtime": 7.7356,
      "eval_samples_per_second": 484.771,
      "eval_steps_per_second": 7.627,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 214.21957397460938,
      "learning_rate": 2.25e-05,
      "loss": 1.4357,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7850666666666667,
      "eval_loss": 1.0253041982650757,
      "eval_runtime": 7.6401,
      "eval_samples_per_second": 490.833,
      "eval_steps_per_second": 7.722,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 362.7677307128906,
      "learning_rate": 1.760472266500396e-05,
      "loss": 0.8316,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8221333333333334,
      "eval_loss": 0.7071218490600586,
      "eval_runtime": 7.6281,
      "eval_samples_per_second": 491.602,
      "eval_steps_per_second": 7.735,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 167.91940307617188,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 0.6323,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8416,
      "eval_loss": 0.6082471609115601,
      "eval_runtime": 7.6124,
      "eval_samples_per_second": 492.617,
      "eval_steps_per_second": 7.75,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 114.81352996826172,
      "learning_rate": 7.500000000000004e-06,
      "loss": 0.5374,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8506666666666667,
      "eval_loss": 0.5484508872032166,
      "eval_runtime": 7.6056,
      "eval_samples_per_second": 493.055,
      "eval_steps_per_second": 7.757,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 190.5194854736328,
      "learning_rate": 3.5093333532153316e-06,
      "loss": 0.4875,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.5485090017318726,
      "eval_runtime": 7.6028,
      "eval_samples_per_second": 493.242,
      "eval_steps_per_second": 7.76,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 412.69415283203125,
      "learning_rate": 9.046106882113753e-07,
      "loss": 0.4666,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.5451526045799255,
      "eval_runtime": 7.6706,
      "eval_samples_per_second": 488.881,
      "eval_steps_per_second": 7.692,
      "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
}