trainer_state.json

{
  "best_metric": 0.9026666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0123/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 37.044673919677734,
      "learning_rate": 0.0004522542485937369,
      "loss": 1.2108,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8384,
      "eval_loss": 0.5310590863227844,
      "eval_runtime": 12.3306,
      "eval_samples_per_second": 304.122,
      "eval_steps_per_second": 4.785,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.532936096191406,
      "learning_rate": 0.00032725424859373687,
      "loss": 0.3527,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.4470193088054657,
      "eval_runtime": 12.416,
      "eval_samples_per_second": 302.029,
      "eval_steps_per_second": 4.752,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 56.66950607299805,
      "learning_rate": 0.00017274575140626317,
      "loss": 0.1648,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.36716118454933167,
      "eval_runtime": 12.3847,
      "eval_samples_per_second": 302.793,
      "eval_steps_per_second": 4.764,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 36.68853759765625,
      "learning_rate": 4.7745751406263163e-05,
      "loss": 0.0617,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.3757179081439972,
      "eval_runtime": 12.3082,
      "eval_samples_per_second": 304.675,
      "eval_steps_per_second": 4.794,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 21.246726989746094,
      "learning_rate": 0.0,
      "loss": 0.0322,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9026666666666666,
      "eval_loss": 0.3705720901489258,
      "eval_runtime": 12.591,
      "eval_samples_per_second": 297.831,
      "eval_steps_per_second": 4.686,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}