trainer_state.json

{
  "best_metric": 0.8754666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0883/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": 84.52495574951172,
      "learning_rate": 8.554359905560886e-05,
      "loss": 3.0889,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6586666666666666,
      "eval_loss": 1.3561450242996216,
      "eval_runtime": 7.4525,
      "eval_samples_per_second": 503.187,
      "eval_steps_per_second": 7.917,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 60.134117126464844,
      "learning_rate": 7.305704108364301e-05,
      "loss": 0.8847,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8192,
      "eval_loss": 0.6024465560913086,
      "eval_runtime": 7.4453,
      "eval_samples_per_second": 503.674,
      "eval_steps_per_second": 7.924,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 38.51520919799805,
      "learning_rate": 5.501344202803415e-05,
      "loss": 0.5251,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8472,
      "eval_loss": 0.5036054253578186,
      "eval_runtime": 7.4422,
      "eval_samples_per_second": 503.886,
      "eval_steps_per_second": 7.928,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 57.51735305786133,
      "learning_rate": 3.498655797196586e-05,
      "loss": 0.4057,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.4636121094226837,
      "eval_runtime": 9.1383,
      "eval_samples_per_second": 410.359,
      "eval_steps_per_second": 6.456,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 55.2630615234375,
      "learning_rate": 1.6942958916356995e-05,
      "loss": 0.3319,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.43960872292518616,
      "eval_runtime": 10.8396,
      "eval_samples_per_second": 345.954,
      "eval_steps_per_second": 5.443,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 43.207908630371094,
      "learning_rate": 4.456400944391144e-06,
      "loss": 0.2937,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.432237446308136,
      "eval_runtime": 7.4475,
      "eval_samples_per_second": 503.527,
      "eval_steps_per_second": 7.922,
      "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
}