trainer_state.json

{
  "best_metric": 0.9165333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0743/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": 48.02005386352539,
      "learning_rate": 0.0002571428571428571,
      "loss": 1.4345,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8530666666666666,
      "eval_loss": 0.4853057265281677,
      "eval_runtime": 7.2693,
      "eval_samples_per_second": 515.866,
      "eval_steps_per_second": 8.116,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 46.01153564453125,
      "learning_rate": 0.00021428571428571427,
      "loss": 0.3602,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8946666666666667,
      "eval_loss": 0.36441704630851746,
      "eval_runtime": 7.2607,
      "eval_samples_per_second": 516.477,
      "eval_steps_per_second": 8.126,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 69.61089324951172,
      "learning_rate": 0.0001714285714285714,
      "loss": 0.2074,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8976,
      "eval_loss": 0.3495977818965912,
      "eval_runtime": 7.182,
      "eval_samples_per_second": 522.14,
      "eval_steps_per_second": 8.215,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 36.26875686645508,
      "learning_rate": 0.00012857142857142855,
      "loss": 0.1208,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9050666666666667,
      "eval_loss": 0.3374248445034027,
      "eval_runtime": 7.1743,
      "eval_samples_per_second": 522.702,
      "eval_steps_per_second": 8.224,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 60.403968811035156,
      "learning_rate": 8.57142857142857e-05,
      "loss": 0.0763,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9125333333333333,
      "eval_loss": 0.34153690934181213,
      "eval_runtime": 7.143,
      "eval_samples_per_second": 524.988,
      "eval_steps_per_second": 8.26,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 75.243896484375,
      "learning_rate": 4.285714285714285e-05,
      "loss": 0.0489,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9165333333333333,
      "eval_loss": 0.33862265944480896,
      "eval_runtime": 7.6261,
      "eval_samples_per_second": 491.735,
      "eval_steps_per_second": 7.737,
      "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
}