trainer_state.json

{
  "best_metric": 0.8768,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0057/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": 94.58206176757812,
      "learning_rate": 8.888888888888889e-05,
      "loss": 2.9559,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6618666666666667,
      "eval_loss": 1.2562921047210693,
      "eval_runtime": 5.4608,
      "eval_samples_per_second": 686.715,
      "eval_steps_per_second": 10.804,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 106.94993591308594,
      "learning_rate": 7.777777777777778e-05,
      "loss": 0.8831,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8053333333333333,
      "eval_loss": 0.6474735140800476,
      "eval_runtime": 5.4217,
      "eval_samples_per_second": 691.663,
      "eval_steps_per_second": 10.882,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 103.95709991455078,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.579,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.5489925742149353,
      "eval_runtime": 5.4224,
      "eval_samples_per_second": 691.576,
      "eval_steps_per_second": 10.881,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 105.58402252197266,
      "learning_rate": 5.555555555555556e-05,
      "loss": 0.4424,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.46961361169815063,
      "eval_runtime": 5.3874,
      "eval_samples_per_second": 696.068,
      "eval_steps_per_second": 10.951,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 78.17851257324219,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3356,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.44756340980529785,
      "eval_runtime": 5.5089,
      "eval_samples_per_second": 680.711,
      "eval_steps_per_second": 10.71,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 58.05252456665039,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.2881,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8661333333333333,
      "eval_loss": 0.4374311566352844,
      "eval_runtime": 5.3964,
      "eval_samples_per_second": 694.914,
      "eval_steps_per_second": 10.933,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 119.95729064941406,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.2501,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8730666666666667,
      "eval_loss": 0.4169938862323761,
      "eval_runtime": 5.4161,
      "eval_samples_per_second": 692.384,
      "eval_steps_per_second": 10.894,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 116.36593627929688,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.2233,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.4171982407569885,
      "eval_runtime": 5.7644,
      "eval_samples_per_second": 650.54,
      "eval_steps_per_second": 10.235,
      "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
}