trainer_state.json

{
  "best_metric": 0.876,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0846/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": 77.06702423095703,
      "learning_rate": 4.752422169756048e-05,
      "loss": 3.6613,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5853333333333334,
      "eval_loss": 3.072692632675171,
      "eval_runtime": 7.5701,
      "eval_samples_per_second": 495.367,
      "eval_steps_per_second": 7.794,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 107.75822448730469,
      "learning_rate": 4.058724504646834e-05,
      "loss": 1.8375,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7557333333333334,
      "eval_loss": 0.9727783799171448,
      "eval_runtime": 7.5136,
      "eval_samples_per_second": 499.093,
      "eval_steps_per_second": 7.852,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 215.87725830078125,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.8122,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8309333333333333,
      "eval_loss": 0.646449089050293,
      "eval_runtime": 7.5328,
      "eval_samples_per_second": 497.822,
      "eval_steps_per_second": 7.832,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 389.4276123046875,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.5879,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8546666666666667,
      "eval_loss": 0.525234043598175,
      "eval_runtime": 7.5429,
      "eval_samples_per_second": 497.155,
      "eval_steps_per_second": 7.822,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 123.77269744873047,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.4895,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4961576461791992,
      "eval_runtime": 7.5266,
      "eval_samples_per_second": 498.234,
      "eval_steps_per_second": 7.839,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 76.33296203613281,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.4472,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.46423813700675964,
      "eval_runtime": 7.4963,
      "eval_samples_per_second": 500.246,
      "eval_steps_per_second": 7.871,
      "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
}