trainer_state.json

{
  "best_metric": 0.9090666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0323/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": 45.963172912597656,
      "learning_rate": 0.0002571428571428571,
      "loss": 1.4613,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.5126265287399292,
      "eval_runtime": 7.9263,
      "eval_samples_per_second": 473.107,
      "eval_steps_per_second": 7.444,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 25.985885620117188,
      "learning_rate": 0.00021428571428571427,
      "loss": 0.4147,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.3841055631637573,
      "eval_runtime": 7.962,
      "eval_samples_per_second": 470.986,
      "eval_steps_per_second": 7.41,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 72.99700164794922,
      "learning_rate": 0.0001714285714285714,
      "loss": 0.2345,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.36671993136405945,
      "eval_runtime": 7.8287,
      "eval_samples_per_second": 479.005,
      "eval_steps_per_second": 7.536,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 73.73075866699219,
      "learning_rate": 0.00012857142857142855,
      "loss": 0.1365,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9013333333333333,
      "eval_loss": 0.331452876329422,
      "eval_runtime": 7.8367,
      "eval_samples_per_second": 478.517,
      "eval_steps_per_second": 7.529,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 41.18482971191406,
      "learning_rate": 8.57142857142857e-05,
      "loss": 0.0832,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9050666666666667,
      "eval_loss": 0.33844229578971863,
      "eval_runtime": 7.8274,
      "eval_samples_per_second": 479.084,
      "eval_steps_per_second": 7.538,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 66.04237365722656,
      "learning_rate": 4.285714285714285e-05,
      "loss": 0.0501,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.34817570447921753,
      "eval_runtime": 7.8054,
      "eval_samples_per_second": 480.435,
      "eval_steps_per_second": 7.559,
      "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
}