trainer_state.json

{
  "best_metric": 0.8658666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0173/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": 102.54977416992188,
      "learning_rate": 7.2e-05,
      "loss": 3.232,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6293333333333333,
      "eval_loss": 1.534888505935669,
      "eval_runtime": 9.4244,
      "eval_samples_per_second": 397.905,
      "eval_steps_per_second": 6.26,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 142.2037353515625,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 0.99,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8008,
      "eval_loss": 0.6795520782470703,
      "eval_runtime": 9.3509,
      "eval_samples_per_second": 401.033,
      "eval_steps_per_second": 6.31,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 132.98947143554688,
      "learning_rate": 3.6e-05,
      "loss": 0.6225,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8466666666666667,
      "eval_loss": 0.5301174521446228,
      "eval_runtime": 9.2103,
      "eval_samples_per_second": 407.153,
      "eval_steps_per_second": 6.406,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 153.4900665283203,
      "learning_rate": 1.8e-05,
      "loss": 0.5064,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.45383352041244507,
      "eval_runtime": 9.0287,
      "eval_samples_per_second": 415.344,
      "eval_steps_per_second": 6.535,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 92.26731872558594,
      "learning_rate": 0.0,
      "loss": 0.4488,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.4424925148487091,
      "eval_runtime": 9.2353,
      "eval_samples_per_second": 406.051,
      "eval_steps_per_second": 6.389,
      "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
}