trainer_state.json

{
  "best_metric": 0.6445333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0284/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": 187.50315856933594,
      "learning_rate": 2.4e-05,
      "loss": 3.8646,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2568,
      "eval_loss": 3.779456853866577,
      "eval_runtime": 8.0695,
      "eval_samples_per_second": 464.714,
      "eval_steps_per_second": 7.312,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 173.8251953125,
      "learning_rate": 1.8e-05,
      "loss": 3.4992,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4752,
      "eval_loss": 3.2124099731445312,
      "eval_runtime": 7.4557,
      "eval_samples_per_second": 502.973,
      "eval_steps_per_second": 7.913,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 144.54368591308594,
      "learning_rate": 1.2e-05,
      "loss": 2.7832,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5698666666666666,
      "eval_loss": 2.440659999847412,
      "eval_runtime": 7.4928,
      "eval_samples_per_second": 500.479,
      "eval_steps_per_second": 7.874,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 200.75082397460938,
      "learning_rate": 6e-06,
      "loss": 2.1416,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6224,
      "eval_loss": 1.9173475503921509,
      "eval_runtime": 7.4566,
      "eval_samples_per_second": 502.913,
      "eval_steps_per_second": 7.912,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 77.96934509277344,
      "learning_rate": 0.0,
      "loss": 1.86,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6445333333333333,
      "eval_loss": 1.866434097290039,
      "eval_runtime": 7.4846,
      "eval_samples_per_second": 501.029,
      "eval_steps_per_second": 7.883,
      "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
}