trainer_state.json

{
  "best_metric": 0.8858666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0989/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": 30.894359588623047,
      "learning_rate": 0.0004,
      "loss": 1.4135,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8069333333333333,
      "eval_loss": 0.6328855752944946,
      "eval_runtime": 8.9304,
      "eval_samples_per_second": 419.913,
      "eval_steps_per_second": 6.607,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 33.403099060058594,
      "learning_rate": 0.0003,
      "loss": 0.4726,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8618666666666667,
      "eval_loss": 0.4408737123012543,
      "eval_runtime": 9.2969,
      "eval_samples_per_second": 403.361,
      "eval_steps_per_second": 6.346,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 28.0541934967041,
      "learning_rate": 0.0002,
      "loss": 0.2748,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.3926353454589844,
      "eval_runtime": 9.2086,
      "eval_samples_per_second": 407.228,
      "eval_steps_per_second": 6.407,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 45.619232177734375,
      "learning_rate": 0.0001,
      "loss": 0.157,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.392044335603714,
      "eval_runtime": 8.8832,
      "eval_samples_per_second": 422.147,
      "eval_steps_per_second": 6.642,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 38.61171340942383,
      "learning_rate": 0.0,
      "loss": 0.0965,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8858666666666667,
      "eval_loss": 0.3921462893486023,
      "eval_runtime": 8.9967,
      "eval_samples_per_second": 416.818,
      "eval_steps_per_second": 6.558,
      "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
}