trainer_state.json

{
  "best_metric": 0.8816,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0845/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": 174.55865478515625,
      "learning_rate": 0.0001,
      "loss": 2.8537,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7189333333333333,
      "eval_loss": 1.0722390413284302,
      "eval_runtime": 8.9922,
      "eval_samples_per_second": 417.03,
      "eval_steps_per_second": 6.561,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 89.96263885498047,
      "learning_rate": 0.0001,
      "loss": 0.8144,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8296,
      "eval_loss": 0.5742604732513428,
      "eval_runtime": 9.3174,
      "eval_samples_per_second": 402.474,
      "eval_steps_per_second": 6.332,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 90.50025177001953,
      "learning_rate": 0.0001,
      "loss": 0.5271,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8477333333333333,
      "eval_loss": 0.5025801062583923,
      "eval_runtime": 9.1163,
      "eval_samples_per_second": 411.351,
      "eval_steps_per_second": 6.472,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 257.1667175292969,
      "learning_rate": 0.0001,
      "loss": 0.3894,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.4274735748767853,
      "eval_runtime": 9.3098,
      "eval_samples_per_second": 402.8,
      "eval_steps_per_second": 6.337,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 137.42127990722656,
      "learning_rate": 0.0001,
      "loss": 0.3098,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.42457032203674316,
      "eval_runtime": 9.3066,
      "eval_samples_per_second": 402.939,
      "eval_steps_per_second": 6.34,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 92.85485076904297,
      "learning_rate": 0.0001,
      "loss": 0.2298,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8816,
      "eval_loss": 0.4004196226596832,
      "eval_runtime": 9.3451,
      "eval_samples_per_second": 401.281,
      "eval_steps_per_second": 6.313,
      "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
}