trainer_state.json

{
  "best_metric": 0.8544,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0508/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": 29.3706111907959,
      "learning_rate": 0.0005,
      "loss": 1.2974,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8133333333333334,
      "eval_loss": 0.6221215724945068,
      "eval_runtime": 12.316,
      "eval_samples_per_second": 304.481,
      "eval_steps_per_second": 4.791,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 39.48615264892578,
      "learning_rate": 0.0005,
      "loss": 0.4273,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8405333333333334,
      "eval_loss": 0.5527553558349609,
      "eval_runtime": 12.9523,
      "eval_samples_per_second": 289.523,
      "eval_steps_per_second": 4.555,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 28.045490264892578,
      "learning_rate": 0.0005,
      "loss": 0.252,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8397333333333333,
      "eval_loss": 0.5832853317260742,
      "eval_runtime": 12.6455,
      "eval_samples_per_second": 296.549,
      "eval_steps_per_second": 4.666,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 19.748994827270508,
      "learning_rate": 0.0005,
      "loss": 0.1781,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8522666666666666,
      "eval_loss": 0.5380047559738159,
      "eval_runtime": 12.3891,
      "eval_samples_per_second": 302.686,
      "eval_steps_per_second": 4.762,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 36.11511993408203,
      "learning_rate": 0.0005,
      "loss": 0.1395,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8525333333333334,
      "eval_loss": 0.5955322980880737,
      "eval_runtime": 12.2256,
      "eval_samples_per_second": 306.733,
      "eval_steps_per_second": 4.826,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 24.674785614013672,
      "learning_rate": 0.0005,
      "loss": 0.1025,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.6214481592178345,
      "eval_runtime": 12.2535,
      "eval_samples_per_second": 306.034,
      "eval_steps_per_second": 4.815,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 31.995697021484375,
      "learning_rate": 0.0005,
      "loss": 0.0888,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8496,
      "eval_loss": 0.6372752785682678,
      "eval_runtime": 12.6811,
      "eval_samples_per_second": 295.717,
      "eval_steps_per_second": 4.653,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 25.495168685913086,
      "learning_rate": 0.0005,
      "loss": 0.0772,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8544,
      "eval_loss": 0.6913248896598816,
      "eval_runtime": 12.2932,
      "eval_samples_per_second": 305.046,
      "eval_steps_per_second": 4.799,
      "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
}