trainer_state.json

{
  "best_metric": 0.896,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0929/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": 38.14113998413086,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.5884,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7722666666666667,
      "eval_loss": 0.8461204171180725,
      "eval_runtime": 13.8316,
      "eval_samples_per_second": 271.119,
      "eval_steps_per_second": 4.266,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 51.603004455566406,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.6056,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8538666666666667,
      "eval_loss": 0.4832752048969269,
      "eval_runtime": 13.7634,
      "eval_samples_per_second": 272.463,
      "eval_steps_per_second": 4.287,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 82.5782470703125,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.3626,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.42033612728118896,
      "eval_runtime": 13.9966,
      "eval_samples_per_second": 267.923,
      "eval_steps_per_second": 4.215,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 264.0753479003906,
      "learning_rate": 5e-05,
      "loss": 0.2562,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8853333333333333,
      "eval_loss": 0.383017361164093,
      "eval_runtime": 13.733,
      "eval_samples_per_second": 273.066,
      "eval_steps_per_second": 4.296,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 63.85730743408203,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.1874,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8893333333333333,
      "eval_loss": 0.37134039402008057,
      "eval_runtime": 13.9275,
      "eval_samples_per_second": 269.251,
      "eval_steps_per_second": 4.236,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 81.89769744873047,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.1518,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.3564470112323761,
      "eval_runtime": 13.7549,
      "eval_samples_per_second": 272.629,
      "eval_steps_per_second": 4.289,
      "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
}