trainer_state.json

{
  "best_metric": 0.9010666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0622/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 68.05856323242188,
      "learning_rate": 0.0004809698831278217,
      "loss": 1.387,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8293333333333334,
      "eval_loss": 0.5948890447616577,
      "eval_runtime": 7.5564,
      "eval_samples_per_second": 496.266,
      "eval_steps_per_second": 7.808,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 102.17045593261719,
      "learning_rate": 0.00042677669529663686,
      "loss": 0.4731,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8469333333333333,
      "eval_loss": 0.5295273065567017,
      "eval_runtime": 7.4246,
      "eval_samples_per_second": 505.08,
      "eval_steps_per_second": 7.947,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 28.552093505859375,
      "learning_rate": 0.0003456708580912725,
      "loss": 0.2744,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.442254900932312,
      "eval_runtime": 7.4193,
      "eval_samples_per_second": 505.436,
      "eval_steps_per_second": 7.952,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 26.617876052856445,
      "learning_rate": 0.00025,
      "loss": 0.1538,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4343659579753876,
      "eval_runtime": 7.4436,
      "eval_samples_per_second": 503.786,
      "eval_steps_per_second": 7.926,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 25.32646369934082,
      "learning_rate": 0.00015432914190872756,
      "loss": 0.085,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8853333333333333,
      "eval_loss": 0.4687633514404297,
      "eval_runtime": 7.4638,
      "eval_samples_per_second": 502.422,
      "eval_steps_per_second": 7.905,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 21.33710479736328,
      "learning_rate": 7.322330470336314e-05,
      "loss": 0.0486,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.408098965883255,
      "eval_runtime": 7.8513,
      "eval_samples_per_second": 477.629,
      "eval_steps_per_second": 7.515,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 28.018333435058594,
      "learning_rate": 1.9030116872178316e-05,
      "loss": 0.0307,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9010666666666667,
      "eval_loss": 0.42140933871269226,
      "eval_runtime": 7.4751,
      "eval_samples_per_second": 501.668,
      "eval_steps_per_second": 7.893,
      "step": 2331
    }
  ],
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}