trainer_state.json

{
  "best_metric": 0.8586666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0264/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 74.59566497802734,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 3.6931,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5072,
      "eval_loss": 3.1311843395233154,
      "eval_runtime": 9.5942,
      "eval_samples_per_second": 390.861,
      "eval_steps_per_second": 6.15,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 161.89517211914062,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.7261,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7677333333333334,
      "eval_loss": 0.9240449070930481,
      "eval_runtime": 9.3474,
      "eval_samples_per_second": 401.18,
      "eval_steps_per_second": 6.312,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 182.5560302734375,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.8012,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8152,
      "eval_loss": 0.6453301310539246,
      "eval_runtime": 9.4281,
      "eval_samples_per_second": 397.746,
      "eval_steps_per_second": 6.258,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 108.04351806640625,
      "learning_rate": 2.5e-05,
      "loss": 0.5911,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8325333333333333,
      "eval_loss": 0.5468270778656006,
      "eval_runtime": 9.564,
      "eval_samples_per_second": 392.097,
      "eval_steps_per_second": 6.169,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 90.23300170898438,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.5094,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8416,
      "eval_loss": 0.509234607219696,
      "eval_runtime": 9.6124,
      "eval_samples_per_second": 390.122,
      "eval_steps_per_second": 6.138,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 93.55977630615234,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.4656,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8469333333333333,
      "eval_loss": 0.5088739991188049,
      "eval_runtime": 9.5852,
      "eval_samples_per_second": 391.229,
      "eval_steps_per_second": 6.155,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 204.49539184570312,
      "learning_rate": 1.9030116872178316e-06,
      "loss": 0.4401,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8536,
      "eval_loss": 0.48828405141830444,
      "eval_runtime": 9.3885,
      "eval_samples_per_second": 399.423,
      "eval_steps_per_second": 6.284,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 146.7166290283203,
      "learning_rate": 0.0,
      "loss": 0.4345,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.478254109621048,
      "eval_runtime": 9.3705,
      "eval_samples_per_second": 400.194,
      "eval_steps_per_second": 6.296,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}