trainer_state.json

{
  "best_metric": 0.8664,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0726/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 94.78150939941406,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.8334,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6962666666666667,
      "eval_loss": 1.0948437452316284,
      "eval_runtime": 16.2452,
      "eval_samples_per_second": 230.838,
      "eval_steps_per_second": 3.632,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 98.2506332397461,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.7985,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.6177684664726257,
      "eval_runtime": 15.4783,
      "eval_samples_per_second": 242.275,
      "eval_steps_per_second": 3.812,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 111.60298919677734,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.537,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8368,
      "eval_loss": 0.5111733675003052,
      "eval_runtime": 15.7953,
      "eval_samples_per_second": 237.412,
      "eval_steps_per_second": 3.735,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 124.0320053100586,
      "learning_rate": 5e-05,
      "loss": 0.4035,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.45756644010543823,
      "eval_runtime": 15.4439,
      "eval_samples_per_second": 242.814,
      "eval_steps_per_second": 3.82,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 65.19634246826172,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.3368,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.4321005344390869,
      "eval_runtime": 15.4337,
      "eval_samples_per_second": 242.975,
      "eval_steps_per_second": 3.823,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 65.79479217529297,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.2938,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.43435850739479065,
      "eval_runtime": 15.4189,
      "eval_samples_per_second": 243.208,
      "eval_steps_per_second": 3.826,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 106.61091613769531,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.2717,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.44640854001045227,
      "eval_runtime": 15.4235,
      "eval_samples_per_second": 243.135,
      "eval_steps_per_second": 3.825,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 70.20097351074219,
      "learning_rate": 0.0,
      "loss": 0.2662,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.43316665291786194,
      "eval_runtime": 15.3323,
      "eval_samples_per_second": 244.582,
      "eval_steps_per_second": 3.848,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}