trainer_state.json

{
  "best_metric": 0.8624,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0202/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 266.8658142089844,
      "learning_rate": 4.752422169756048e-05,
      "loss": 3.5868,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6189333333333333,
      "eval_loss": 2.8215999603271484,
      "eval_runtime": 12.709,
      "eval_samples_per_second": 295.068,
      "eval_steps_per_second": 4.642,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 327.2888488769531,
      "learning_rate": 4.058724504646834e-05,
      "loss": 1.6073,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7936,
      "eval_loss": 0.8350662589073181,
      "eval_runtime": 12.6952,
      "eval_samples_per_second": 295.388,
      "eval_steps_per_second": 4.647,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 119.57669067382812,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.6824,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8416,
      "eval_loss": 0.5731309652328491,
      "eval_runtime": 13.0164,
      "eval_samples_per_second": 288.097,
      "eval_steps_per_second": 4.533,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 230.3582763671875,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.4992,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8530666666666666,
      "eval_loss": 0.4975144565105438,
      "eval_runtime": 12.5571,
      "eval_samples_per_second": 298.635,
      "eval_steps_per_second": 4.699,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 104.05943298339844,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.4282,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.4668174982070923,
      "eval_runtime": 12.9852,
      "eval_samples_per_second": 288.791,
      "eval_steps_per_second": 4.544,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 276.62615966796875,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.3908,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.46305084228515625,
      "eval_runtime": 12.5794,
      "eval_samples_per_second": 298.107,
      "eval_steps_per_second": 4.69,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}