trainer_state.json

{
  "best_metric": 0.8794666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0176/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": 153.32424926757812,
      "learning_rate": 5e-05,
      "loss": 3.603,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5424,
      "eval_loss": 2.721700429916382,
      "eval_runtime": 12.7761,
      "eval_samples_per_second": 293.516,
      "eval_steps_per_second": 4.618,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 132.0540008544922,
      "learning_rate": 5e-05,
      "loss": 1.4729,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7706666666666667,
      "eval_loss": 0.8357215523719788,
      "eval_runtime": 13.3225,
      "eval_samples_per_second": 281.479,
      "eval_steps_per_second": 4.429,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 88.62503051757812,
      "learning_rate": 5e-05,
      "loss": 0.6969,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8317333333333333,
      "eval_loss": 0.5686293244361877,
      "eval_runtime": 12.9802,
      "eval_samples_per_second": 288.902,
      "eval_steps_per_second": 4.545,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 69.73323822021484,
      "learning_rate": 5e-05,
      "loss": 0.4978,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.4945754408836365,
      "eval_runtime": 13.1067,
      "eval_samples_per_second": 286.112,
      "eval_steps_per_second": 4.501,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 148.49737548828125,
      "learning_rate": 5e-05,
      "loss": 0.3866,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.4511866867542267,
      "eval_runtime": 13.4847,
      "eval_samples_per_second": 278.092,
      "eval_steps_per_second": 4.375,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 134.58824157714844,
      "learning_rate": 5e-05,
      "loss": 0.3067,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.40360721945762634,
      "eval_runtime": 13.4696,
      "eval_samples_per_second": 278.404,
      "eval_steps_per_second": 4.38,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 303.9842224121094,
      "learning_rate": 5e-05,
      "loss": 0.2503,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8730666666666667,
      "eval_loss": 0.40265795588493347,
      "eval_runtime": 13.5282,
      "eval_samples_per_second": 277.2,
      "eval_steps_per_second": 4.361,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 105.24630737304688,
      "learning_rate": 5e-05,
      "loss": 0.2091,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.3984690010547638,
      "eval_runtime": 12.9465,
      "eval_samples_per_second": 289.654,
      "eval_steps_per_second": 4.557,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}