trainer_state.json

{
  "best_metric": 0.8818666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0936/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": 123.5322265625,
      "learning_rate": 7.875e-05,
      "loss": 2.935,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7186666666666667,
      "eval_loss": 1.11838698387146,
      "eval_runtime": 12.8297,
      "eval_samples_per_second": 292.29,
      "eval_steps_per_second": 4.599,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 203.44822692871094,
      "learning_rate": 6.75e-05,
      "loss": 0.7703,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5366340279579163,
      "eval_runtime": 12.6404,
      "eval_samples_per_second": 296.668,
      "eval_steps_per_second": 4.668,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 92.3979721069336,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 0.4725,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.46114933490753174,
      "eval_runtime": 12.8187,
      "eval_samples_per_second": 292.541,
      "eval_steps_per_second": 4.603,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 94.81336975097656,
      "learning_rate": 4.5e-05,
      "loss": 0.3527,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.41661667823791504,
      "eval_runtime": 12.8087,
      "eval_samples_per_second": 292.77,
      "eval_steps_per_second": 4.606,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 160.11831665039062,
      "learning_rate": 3.375e-05,
      "loss": 0.2777,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.41281530261039734,
      "eval_runtime": 12.5821,
      "eval_samples_per_second": 298.043,
      "eval_steps_per_second": 4.689,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 58.51556396484375,
      "learning_rate": 2.25e-05,
      "loss": 0.2319,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.3918461501598358,
      "eval_runtime": 12.6296,
      "eval_samples_per_second": 296.923,
      "eval_steps_per_second": 4.672,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 99.05912780761719,
      "learning_rate": 1.125e-05,
      "loss": 0.1958,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8773333333333333,
      "eval_loss": 0.3983180522918701,
      "eval_runtime": 12.633,
      "eval_samples_per_second": 296.841,
      "eval_steps_per_second": 4.67,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 171.2135772705078,
      "learning_rate": 0.0,
      "loss": 0.1819,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.3878120481967926,
      "eval_runtime": 12.625,
      "eval_samples_per_second": 297.03,
      "eval_steps_per_second": 4.673,
      "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
}