trainer_state.json

{
  "best_metric": 0.8976,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0029/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": 45.324214935302734,
      "learning_rate": 0.0004849231551964771,
      "loss": 1.4214,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.6353247761726379,
      "eval_runtime": 9.1769,
      "eval_samples_per_second": 408.635,
      "eval_steps_per_second": 6.429,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.407304763793945,
      "learning_rate": 0.0004415111107797445,
      "loss": 0.5013,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8416,
      "eval_loss": 0.55496746301651,
      "eval_runtime": 9.0189,
      "eval_samples_per_second": 415.792,
      "eval_steps_per_second": 6.542,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 16.43842315673828,
      "learning_rate": 0.000375,
      "loss": 0.2982,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.5013601183891296,
      "eval_runtime": 9.034,
      "eval_samples_per_second": 415.097,
      "eval_steps_per_second": 6.531,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 30.339820861816406,
      "learning_rate": 0.00029341204441673266,
      "loss": 0.1787,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.496133416891098,
      "eval_runtime": 9.4506,
      "eval_samples_per_second": 396.799,
      "eval_steps_per_second": 6.243,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.85334014892578,
      "learning_rate": 0.00020658795558326743,
      "loss": 0.11,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.48969927430152893,
      "eval_runtime": 9.0257,
      "eval_samples_per_second": 415.48,
      "eval_steps_per_second": 6.537,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 21.973466873168945,
      "learning_rate": 0.00012500000000000006,
      "loss": 0.0622,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.46728047728538513,
      "eval_runtime": 9.2554,
      "eval_samples_per_second": 405.168,
      "eval_steps_per_second": 6.375,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 22.861583709716797,
      "learning_rate": 5.848888922025553e-05,
      "loss": 0.0333,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.4697713553905487,
      "eval_runtime": 9.0861,
      "eval_samples_per_second": 412.72,
      "eval_steps_per_second": 6.493,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 31.28031349182129,
      "learning_rate": 1.5076844803522921e-05,
      "loss": 0.0301,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8976,
      "eval_loss": 0.4572010040283203,
      "eval_runtime": 9.3725,
      "eval_samples_per_second": 400.105,
      "eval_steps_per_second": 6.295,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}