trainer_state.json

{
  "best_metric": 0.8402666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0230/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": 114.92549133300781,
      "learning_rate": 4.752422169756048e-05,
      "loss": 3.7102,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4789333333333333,
      "eval_loss": 3.1450417041778564,
      "eval_runtime": 8.111,
      "eval_samples_per_second": 462.336,
      "eval_steps_per_second": 7.274,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 145.68724060058594,
      "learning_rate": 4.058724504646834e-05,
      "loss": 1.9109,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7258666666666667,
      "eval_loss": 1.0468553304672241,
      "eval_runtime": 8.1212,
      "eval_samples_per_second": 461.756,
      "eval_steps_per_second": 7.265,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 76.70687866210938,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.8943,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7949333333333334,
      "eval_loss": 0.7348412871360779,
      "eval_runtime": 8.0669,
      "eval_samples_per_second": 464.862,
      "eval_steps_per_second": 7.314,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 149.61343383789062,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.6818,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.82,
      "eval_loss": 0.6111552119255066,
      "eval_runtime": 8.0832,
      "eval_samples_per_second": 463.926,
      "eval_steps_per_second": 7.299,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 185.0524444580078,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.5962,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8312,
      "eval_loss": 0.5720964074134827,
      "eval_runtime": 8.149,
      "eval_samples_per_second": 460.179,
      "eval_steps_per_second": 7.24,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 133.14625549316406,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.5487,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5476441383361816,
      "eval_runtime": 8.5597,
      "eval_samples_per_second": 438.099,
      "eval_steps_per_second": 6.893,
      "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
}