trainer_state.json

{
  "best_metric": 0.8229333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0502/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": 129.3270721435547,
      "learning_rate": 4e-05,
      "loss": 3.641,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5549333333333333,
      "eval_loss": 2.9843344688415527,
      "eval_runtime": 7.7866,
      "eval_samples_per_second": 481.597,
      "eval_steps_per_second": 7.577,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 280.9588623046875,
      "learning_rate": 3e-05,
      "loss": 1.8503,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7365333333333334,
      "eval_loss": 1.110457181930542,
      "eval_runtime": 7.6634,
      "eval_samples_per_second": 489.341,
      "eval_steps_per_second": 7.699,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 155.38815307617188,
      "learning_rate": 2e-05,
      "loss": 0.9233,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7994666666666667,
      "eval_loss": 0.7700660228729248,
      "eval_runtime": 7.6174,
      "eval_samples_per_second": 492.297,
      "eval_steps_per_second": 7.745,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 120.37042999267578,
      "learning_rate": 1e-05,
      "loss": 0.7016,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8197333333333333,
      "eval_loss": 0.6470049619674683,
      "eval_runtime": 7.5909,
      "eval_samples_per_second": 494.013,
      "eval_steps_per_second": 7.772,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 179.9027099609375,
      "learning_rate": 0.0,
      "loss": 0.6223,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8229333333333333,
      "eval_loss": 0.6286609172821045,
      "eval_runtime": 7.5607,
      "eval_samples_per_second": 495.987,
      "eval_steps_per_second": 7.804,
      "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
}