trainer_state.json

{
  "best_metric": 0.8933333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0269/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": 41.495323181152344,
      "learning_rate": 0.0004166666666666667,
      "loss": 1.3859,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8288,
      "eval_loss": 0.5539724230766296,
      "eval_runtime": 15.2924,
      "eval_samples_per_second": 245.219,
      "eval_steps_per_second": 3.858,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 44.347408294677734,
      "learning_rate": 0.0003333333333333333,
      "loss": 0.4435,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8714666666666666,
      "eval_loss": 0.4165392518043518,
      "eval_runtime": 15.2893,
      "eval_samples_per_second": 245.269,
      "eval_steps_per_second": 3.859,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 34.140743255615234,
      "learning_rate": 0.00025,
      "loss": 0.2599,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.40256330370903015,
      "eval_runtime": 15.4511,
      "eval_samples_per_second": 242.701,
      "eval_steps_per_second": 3.819,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 38.88068389892578,
      "learning_rate": 0.00016666666666666666,
      "loss": 0.1499,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8858666666666667,
      "eval_loss": 0.43554678559303284,
      "eval_runtime": 15.254,
      "eval_samples_per_second": 245.837,
      "eval_steps_per_second": 3.868,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 27.989450454711914,
      "learning_rate": 8.333333333333333e-05,
      "loss": 0.0838,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8914666666666666,
      "eval_loss": 0.4012109935283661,
      "eval_runtime": 15.2507,
      "eval_samples_per_second": 245.891,
      "eval_steps_per_second": 3.869,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 28.045108795166016,
      "learning_rate": 0.0,
      "loss": 0.0503,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8933333333333333,
      "eval_loss": 0.41423508524894714,
      "eval_runtime": 15.2947,
      "eval_samples_per_second": 245.184,
      "eval_steps_per_second": 3.858,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}