trainer_state.json

{
  "best_metric": 0.9208,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0432/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 35.23207092285156,
      "learning_rate": 0.0004,
      "loss": 1.1927,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.864,
      "eval_loss": 0.45179441571235657,
      "eval_runtime": 14.2307,
      "eval_samples_per_second": 263.514,
      "eval_steps_per_second": 4.146,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 35.69432830810547,
      "learning_rate": 0.0003,
      "loss": 0.3036,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.38089701533317566,
      "eval_runtime": 14.1347,
      "eval_samples_per_second": 265.305,
      "eval_steps_per_second": 4.174,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 31.058170318603516,
      "learning_rate": 0.0002,
      "loss": 0.1574,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9034666666666666,
      "eval_loss": 0.33191901445388794,
      "eval_runtime": 14.2903,
      "eval_samples_per_second": 262.416,
      "eval_steps_per_second": 4.129,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 30.20444107055664,
      "learning_rate": 0.0001,
      "loss": 0.0649,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9117333333333333,
      "eval_loss": 0.34749388694763184,
      "eval_runtime": 14.093,
      "eval_samples_per_second": 266.09,
      "eval_steps_per_second": 4.186,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 41.45025634765625,
      "learning_rate": 0.0,
      "loss": 0.035,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9208,
      "eval_loss": 0.29334700107574463,
      "eval_runtime": 14.0684,
      "eval_samples_per_second": 266.555,
      "eval_steps_per_second": 4.194,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}