trainer_state.json

{
  "best_metric": 0.9184,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0573/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 58.874351501464844,
      "learning_rate": 0.0002625,
      "loss": 1.4992,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8544,
      "eval_loss": 0.4956533908843994,
      "eval_runtime": 7.171,
      "eval_samples_per_second": 522.94,
      "eval_steps_per_second": 8.228,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 39.330650329589844,
      "learning_rate": 0.000225,
      "loss": 0.38,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8898666666666667,
      "eval_loss": 0.35987281799316406,
      "eval_runtime": 7.1083,
      "eval_samples_per_second": 527.556,
      "eval_steps_per_second": 8.3,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 29.70055389404297,
      "learning_rate": 0.00018749999999999998,
      "loss": 0.2221,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.3579879701137543,
      "eval_runtime": 7.0609,
      "eval_samples_per_second": 531.095,
      "eval_steps_per_second": 8.356,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 42.42390823364258,
      "learning_rate": 0.00015,
      "loss": 0.137,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9066666666666666,
      "eval_loss": 0.3296903371810913,
      "eval_runtime": 7.2352,
      "eval_samples_per_second": 518.298,
      "eval_steps_per_second": 8.155,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 52.43452072143555,
      "learning_rate": 0.0001125,
      "loss": 0.095,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.3313167691230774,
      "eval_runtime": 7.089,
      "eval_samples_per_second": 528.99,
      "eval_steps_per_second": 8.323,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 36.34873580932617,
      "learning_rate": 7.5e-05,
      "loss": 0.056,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9154666666666667,
      "eval_loss": 0.3340315520763397,
      "eval_runtime": 7.3154,
      "eval_samples_per_second": 512.616,
      "eval_steps_per_second": 8.065,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 65.05775451660156,
      "learning_rate": 3.75e-05,
      "loss": 0.0381,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9184,
      "eval_loss": 0.31702953577041626,
      "eval_runtime": 7.2807,
      "eval_samples_per_second": 515.062,
      "eval_steps_per_second": 8.104,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}