trainer_state.json

{
  "best_metric": 0.8909333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0767/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": 110.37844848632812,
      "learning_rate": 7e-05,
      "loss": 3.1508,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7018666666666666,
      "eval_loss": 1.3323915004730225,
      "eval_runtime": 8.5779,
      "eval_samples_per_second": 437.169,
      "eval_steps_per_second": 6.878,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 131.10369873046875,
      "learning_rate": 7e-05,
      "loss": 0.8581,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8405333333333334,
      "eval_loss": 0.567375123500824,
      "eval_runtime": 8.5223,
      "eval_samples_per_second": 440.019,
      "eval_steps_per_second": 6.923,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 207.48483276367188,
      "learning_rate": 7e-05,
      "loss": 0.4832,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.4532061219215393,
      "eval_runtime": 9.225,
      "eval_samples_per_second": 406.504,
      "eval_steps_per_second": 6.396,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 56.98298645019531,
      "learning_rate": 7e-05,
      "loss": 0.3069,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.4230201244354248,
      "eval_runtime": 8.6304,
      "eval_samples_per_second": 434.51,
      "eval_steps_per_second": 6.836,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 69.74508666992188,
      "learning_rate": 7e-05,
      "loss": 0.2113,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.39584019780158997,
      "eval_runtime": 8.6424,
      "eval_samples_per_second": 433.908,
      "eval_steps_per_second": 6.827,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 159.8357696533203,
      "learning_rate": 7e-05,
      "loss": 0.1497,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.3842519223690033,
      "eval_runtime": 8.7364,
      "eval_samples_per_second": 429.238,
      "eval_steps_per_second": 6.753,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}