trainer_state.json

{
  "best_metric": 0.8861333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0152/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": 111.4142074584961,
      "learning_rate": 5e-05,
      "loss": 3.6715,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4882666666666667,
      "eval_loss": 3.1333963871002197,
      "eval_runtime": 7.7403,
      "eval_samples_per_second": 484.475,
      "eval_steps_per_second": 7.622,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 81.93476104736328,
      "learning_rate": 5e-05,
      "loss": 1.8773,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7506666666666667,
      "eval_loss": 1.0524919033050537,
      "eval_runtime": 7.7191,
      "eval_samples_per_second": 485.809,
      "eval_steps_per_second": 7.643,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 86.99642181396484,
      "learning_rate": 5e-05,
      "loss": 0.835,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8109333333333333,
      "eval_loss": 0.6627472639083862,
      "eval_runtime": 7.7519,
      "eval_samples_per_second": 483.749,
      "eval_steps_per_second": 7.611,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 188.76773071289062,
      "learning_rate": 5e-05,
      "loss": 0.594,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8424,
      "eval_loss": 0.5403102040290833,
      "eval_runtime": 7.7081,
      "eval_samples_per_second": 486.501,
      "eval_steps_per_second": 7.654,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 90.717041015625,
      "learning_rate": 5e-05,
      "loss": 0.4692,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.46931809186935425,
      "eval_runtime": 7.7391,
      "eval_samples_per_second": 484.555,
      "eval_steps_per_second": 7.624,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 89.55489349365234,
      "learning_rate": 5e-05,
      "loss": 0.3966,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8618666666666667,
      "eval_loss": 0.46468842029571533,
      "eval_runtime": 7.7229,
      "eval_samples_per_second": 485.568,
      "eval_steps_per_second": 7.64,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 130.94464111328125,
      "learning_rate": 5e-05,
      "loss": 0.3278,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.41801875829696655,
      "eval_runtime": 7.7517,
      "eval_samples_per_second": 483.763,
      "eval_steps_per_second": 7.611,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 107.10636901855469,
      "learning_rate": 5e-05,
      "loss": 0.2969,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.38875311613082886,
      "eval_runtime": 7.6905,
      "eval_samples_per_second": 487.616,
      "eval_steps_per_second": 7.672,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}