trainer_state.json

{
  "best_metric": 0.912,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0101/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": 32.01685333251953,
      "learning_rate": 0.0004375,
      "loss": 1.3926,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8477333333333333,
      "eval_loss": 0.5304312705993652,
      "eval_runtime": 7.7326,
      "eval_samples_per_second": 484.957,
      "eval_steps_per_second": 7.63,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.12260627746582,
      "learning_rate": 0.000375,
      "loss": 0.4432,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.45705509185791016,
      "eval_runtime": 7.6134,
      "eval_samples_per_second": 492.555,
      "eval_steps_per_second": 7.75,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 38.68257141113281,
      "learning_rate": 0.0003125,
      "loss": 0.2522,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8877333333333334,
      "eval_loss": 0.4050454795360565,
      "eval_runtime": 7.5951,
      "eval_samples_per_second": 493.74,
      "eval_steps_per_second": 7.768,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 54.05279541015625,
      "learning_rate": 0.00025,
      "loss": 0.1569,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.4059542715549469,
      "eval_runtime": 7.5863,
      "eval_samples_per_second": 494.313,
      "eval_steps_per_second": 7.777,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 25.94581413269043,
      "learning_rate": 0.0001875,
      "loss": 0.0914,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.42724981904029846,
      "eval_runtime": 7.6124,
      "eval_samples_per_second": 492.616,
      "eval_steps_per_second": 7.75,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 32.35019302368164,
      "learning_rate": 0.000125,
      "loss": 0.054,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9016,
      "eval_loss": 0.4181336462497711,
      "eval_runtime": 7.6555,
      "eval_samples_per_second": 489.843,
      "eval_steps_per_second": 7.707,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 42.78337860107422,
      "learning_rate": 6.25e-05,
      "loss": 0.0346,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9085333333333333,
      "eval_loss": 0.37606021761894226,
      "eval_runtime": 7.7496,
      "eval_samples_per_second": 483.898,
      "eval_steps_per_second": 7.613,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 41.785030364990234,
      "learning_rate": 0.0,
      "loss": 0.0216,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.912,
      "eval_loss": 0.38377663493156433,
      "eval_runtime": 7.6558,
      "eval_samples_per_second": 489.823,
      "eval_steps_per_second": 7.707,
      "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
}