trainer_state.json

{
  "best_metric": 0.8754666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0207/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": 37.32371520996094,
      "learning_rate": 0.0005,
      "loss": 1.292,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8210666666666666,
      "eval_loss": 0.6035176515579224,
      "eval_runtime": 7.2527,
      "eval_samples_per_second": 517.047,
      "eval_steps_per_second": 8.135,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.09795570373535,
      "learning_rate": 0.0005,
      "loss": 0.4458,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.5223435163497925,
      "eval_runtime": 7.1486,
      "eval_samples_per_second": 524.577,
      "eval_steps_per_second": 8.253,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 26.49254035949707,
      "learning_rate": 0.0005,
      "loss": 0.2792,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.4586404263973236,
      "eval_runtime": 7.7144,
      "eval_samples_per_second": 486.106,
      "eval_steps_per_second": 7.648,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 27.87540054321289,
      "learning_rate": 0.0005,
      "loss": 0.201,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.864,
      "eval_loss": 0.5108115673065186,
      "eval_runtime": 7.1266,
      "eval_samples_per_second": 526.196,
      "eval_steps_per_second": 8.279,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 32.68070983886719,
      "learning_rate": 0.0005,
      "loss": 0.1509,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8701333333333333,
      "eval_loss": 0.5023248195648193,
      "eval_runtime": 7.1313,
      "eval_samples_per_second": 525.85,
      "eval_steps_per_second": 8.273,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 20.926227569580078,
      "learning_rate": 0.0005,
      "loss": 0.1219,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.49463677406311035,
      "eval_runtime": 7.1484,
      "eval_samples_per_second": 524.595,
      "eval_steps_per_second": 8.254,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}