trainer_state.json

{
  "best_metric": 0.8978666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0934/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": 48.3433952331543,
      "learning_rate": 0.0003,
      "loss": 1.4499,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8373333333333334,
      "eval_loss": 0.5317114591598511,
      "eval_runtime": 14.7118,
      "eval_samples_per_second": 254.898,
      "eval_steps_per_second": 4.01,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 38.0843391418457,
      "learning_rate": 0.0003,
      "loss": 0.4073,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.4175547659397125,
      "eval_runtime": 14.8082,
      "eval_samples_per_second": 253.239,
      "eval_steps_per_second": 3.984,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 37.382659912109375,
      "learning_rate": 0.0003,
      "loss": 0.2631,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.37204602360725403,
      "eval_runtime": 14.1979,
      "eval_samples_per_second": 264.123,
      "eval_steps_per_second": 4.156,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 26.560388565063477,
      "learning_rate": 0.0003,
      "loss": 0.1885,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.4156035780906677,
      "eval_runtime": 14.6404,
      "eval_samples_per_second": 256.141,
      "eval_steps_per_second": 4.03,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 39.094173431396484,
      "learning_rate": 0.0003,
      "loss": 0.1414,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.39890623092651367,
      "eval_runtime": 14.6609,
      "eval_samples_per_second": 255.782,
      "eval_steps_per_second": 4.024,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 48.088134765625,
      "learning_rate": 0.0003,
      "loss": 0.1125,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8968,
      "eval_loss": 0.410978764295578,
      "eval_runtime": 14.7422,
      "eval_samples_per_second": 254.373,
      "eval_steps_per_second": 4.002,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 41.508575439453125,
      "learning_rate": 0.0003,
      "loss": 0.0826,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8978666666666667,
      "eval_loss": 0.43777716159820557,
      "eval_runtime": 15.7313,
      "eval_samples_per_second": 238.379,
      "eval_steps_per_second": 3.75,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}