trainer_state.json

{
  "best_metric": 0.9005333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0676/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": 29.85410499572754,
      "learning_rate": 0.000288581929876693,
      "loss": 1.5427,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5437477827072144,
      "eval_runtime": 14.38,
      "eval_samples_per_second": 260.778,
      "eval_steps_per_second": 4.103,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 58.30100631713867,
      "learning_rate": 0.00025606601717798207,
      "loss": 0.4352,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.864,
      "eval_loss": 0.4501575529575348,
      "eval_runtime": 14.2058,
      "eval_samples_per_second": 263.977,
      "eval_steps_per_second": 4.153,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 37.62248229980469,
      "learning_rate": 0.00020740251485476345,
      "loss": 0.2288,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.41869544982910156,
      "eval_runtime": 14.1039,
      "eval_samples_per_second": 265.884,
      "eval_steps_per_second": 4.183,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 36.39154052734375,
      "learning_rate": 0.00015,
      "loss": 0.1189,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.4265887439250946,
      "eval_runtime": 14.0277,
      "eval_samples_per_second": 267.328,
      "eval_steps_per_second": 4.206,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 38.74934768676758,
      "learning_rate": 9.259748514523653e-05,
      "loss": 0.0572,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.4206765294075012,
      "eval_runtime": 14.1168,
      "eval_samples_per_second": 265.64,
      "eval_steps_per_second": 4.179,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 27.660612106323242,
      "learning_rate": 4.3933982822017876e-05,
      "loss": 0.0307,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9005333333333333,
      "eval_loss": 0.4146912693977356,
      "eval_runtime": 14.1513,
      "eval_samples_per_second": 264.994,
      "eval_steps_per_second": 4.169,
      "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
}