trainer_state.json

{
  "best_metric": 0.8864,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0798/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": 48.65329360961914,
      "learning_rate": 0.0002713525491562421,
      "loss": 1.4836,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5278244614601135,
      "eval_runtime": 7.2605,
      "eval_samples_per_second": 516.497,
      "eval_steps_per_second": 8.126,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 37.58403778076172,
      "learning_rate": 0.0001963525491562421,
      "loss": 0.393,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.435501366853714,
      "eval_runtime": 7.2719,
      "eval_samples_per_second": 515.682,
      "eval_steps_per_second": 8.113,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 52.55616760253906,
      "learning_rate": 0.0001036474508437579,
      "loss": 0.1882,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.4147579073905945,
      "eval_runtime": 7.2201,
      "eval_samples_per_second": 519.383,
      "eval_steps_per_second": 8.172,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 63.16933822631836,
      "learning_rate": 2.8647450843757897e-05,
      "loss": 0.08,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.4034328758716583,
      "eval_runtime": 7.2751,
      "eval_samples_per_second": 515.457,
      "eval_steps_per_second": 8.11,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 29.0805721282959,
      "learning_rate": 0.0,
      "loss": 0.0441,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.4126240015029907,
      "eval_runtime": 7.2319,
      "eval_samples_per_second": 518.534,
      "eval_steps_per_second": 8.158,
      "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
}