trainer_state.json

{
  "best_metric": 0.8861333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0302/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 22.38751983642578,
      "learning_rate": 0.00047524221697560476,
      "loss": 1.4234,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8088,
      "eval_loss": 0.6131624579429626,
      "eval_runtime": 8.8896,
      "eval_samples_per_second": 421.843,
      "eval_steps_per_second": 6.637,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 24.77532196044922,
      "learning_rate": 0.0004058724504646834,
      "loss": 0.4835,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.836,
      "eval_loss": 0.5268007516860962,
      "eval_runtime": 8.5824,
      "eval_samples_per_second": 436.943,
      "eval_steps_per_second": 6.875,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 36.61915588378906,
      "learning_rate": 0.0003056302334890786,
      "loss": 0.2865,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8578666666666667,
      "eval_loss": 0.47999781370162964,
      "eval_runtime": 8.5666,
      "eval_samples_per_second": 437.744,
      "eval_steps_per_second": 6.887,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 43.060001373291016,
      "learning_rate": 0.00019436976651092142,
      "loss": 0.1594,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.4686793386936188,
      "eval_runtime": 8.538,
      "eval_samples_per_second": 439.213,
      "eval_steps_per_second": 6.91,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 29.375308990478516,
      "learning_rate": 9.412754953531663e-05,
      "loss": 0.0853,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.446473628282547,
      "eval_runtime": 8.4497,
      "eval_samples_per_second": 443.8,
      "eval_steps_per_second": 6.982,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 23.510555267333984,
      "learning_rate": 2.4757783024395242e-05,
      "loss": 0.049,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.4372210204601288,
      "eval_runtime": 8.5548,
      "eval_samples_per_second": 438.352,
      "eval_steps_per_second": 6.897,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}