trainer_state.json

{
  "best_metric": 0.8813333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0211/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": 130.8785400390625,
      "learning_rate": 8.657457896300791e-05,
      "loss": 2.976,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6885333333333333,
      "eval_loss": 1.2231577634811401,
      "eval_runtime": 15.4598,
      "eval_samples_per_second": 242.564,
      "eval_steps_per_second": 3.816,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 123.6002197265625,
      "learning_rate": 7.681980515339464e-05,
      "loss": 0.8468,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8125333333333333,
      "eval_loss": 0.6062451601028442,
      "eval_runtime": 15.5517,
      "eval_samples_per_second": 241.132,
      "eval_steps_per_second": 3.794,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 267.19049072265625,
      "learning_rate": 6.222075445642905e-05,
      "loss": 0.537,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8541333333333333,
      "eval_loss": 0.4882449805736542,
      "eval_runtime": 15.5807,
      "eval_samples_per_second": 240.683,
      "eval_steps_per_second": 3.787,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 81.14849853515625,
      "learning_rate": 4.5e-05,
      "loss": 0.4093,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.43807950615882874,
      "eval_runtime": 15.5913,
      "eval_samples_per_second": 240.519,
      "eval_steps_per_second": 3.784,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 60.63694381713867,
      "learning_rate": 2.7779245543570963e-05,
      "loss": 0.3381,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8664,
      "eval_loss": 0.42922207713127136,
      "eval_runtime": 15.7739,
      "eval_samples_per_second": 237.735,
      "eval_steps_per_second": 3.74,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 61.526058197021484,
      "learning_rate": 1.3180194846605365e-05,
      "loss": 0.2975,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.384143203496933,
      "eval_runtime": 15.4366,
      "eval_samples_per_second": 242.928,
      "eval_steps_per_second": 3.822,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}