trainer_state.json

{
  "best_metric": 0.8098666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0543/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": 80.91219329833984,
      "learning_rate": 4e-05,
      "loss": 3.6401,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5605333333333333,
      "eval_loss": 3.195685386657715,
      "eval_runtime": 14.2944,
      "eval_samples_per_second": 262.341,
      "eval_steps_per_second": 4.127,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 436.80511474609375,
      "learning_rate": 3e-05,
      "loss": 2.0882,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7261333333333333,
      "eval_loss": 1.267958641052246,
      "eval_runtime": 14.3376,
      "eval_samples_per_second": 261.551,
      "eval_steps_per_second": 4.115,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 164.52565002441406,
      "learning_rate": 2e-05,
      "loss": 0.9919,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7957333333333333,
      "eval_loss": 0.807950496673584,
      "eval_runtime": 14.2743,
      "eval_samples_per_second": 262.709,
      "eval_steps_per_second": 4.133,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 241.04151916503906,
      "learning_rate": 1e-05,
      "loss": 0.7358,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8053333333333333,
      "eval_loss": 0.6993881464004517,
      "eval_runtime": 14.2631,
      "eval_samples_per_second": 262.916,
      "eval_steps_per_second": 4.137,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 292.1971130371094,
      "learning_rate": 0.0,
      "loss": 0.6396,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8098666666666666,
      "eval_loss": 0.663183331489563,
      "eval_runtime": 14.3027,
      "eval_samples_per_second": 262.187,
      "eval_steps_per_second": 4.125,
      "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
}