trainer_state.json

{
  "best_metric": 0.8752,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0762/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 126.2350082397461,
      "learning_rate": 3e-05,
      "loss": 3.8525,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3258666666666667,
      "eval_loss": 3.7516837120056152,
      "eval_runtime": 13.16,
      "eval_samples_per_second": 284.954,
      "eval_steps_per_second": 4.483,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 115.85437774658203,
      "learning_rate": 3e-05,
      "loss": 3.3057,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5368,
      "eval_loss": 2.760030746459961,
      "eval_runtime": 13.1567,
      "eval_samples_per_second": 285.026,
      "eval_steps_per_second": 4.484,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 320.0165100097656,
      "learning_rate": 3e-05,
      "loss": 1.8488,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7096,
      "eval_loss": 1.2732479572296143,
      "eval_runtime": 13.2512,
      "eval_samples_per_second": 282.993,
      "eval_steps_per_second": 4.452,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 126.75240325927734,
      "learning_rate": 3e-05,
      "loss": 0.9918,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7837333333333333,
      "eval_loss": 0.7941842079162598,
      "eval_runtime": 13.2213,
      "eval_samples_per_second": 283.634,
      "eval_steps_per_second": 4.463,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 84.89559936523438,
      "learning_rate": 3e-05,
      "loss": 0.694,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.6212171316146851,
      "eval_runtime": 13.0968,
      "eval_samples_per_second": 286.329,
      "eval_steps_per_second": 4.505,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 201.0231475830078,
      "learning_rate": 3e-05,
      "loss": 0.5567,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8394666666666667,
      "eval_loss": 0.5455034375190735,
      "eval_runtime": 13.0909,
      "eval_samples_per_second": 286.459,
      "eval_steps_per_second": 4.507,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 174.41970825195312,
      "learning_rate": 3e-05,
      "loss": 0.4854,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.5061615109443665,
      "eval_runtime": 13.2307,
      "eval_samples_per_second": 283.432,
      "eval_steps_per_second": 4.459,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 63.59780502319336,
      "learning_rate": 3e-05,
      "loss": 0.4205,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.856,
      "eval_loss": 0.47387877106666565,
      "eval_runtime": 13.0996,
      "eval_samples_per_second": 286.267,
      "eval_steps_per_second": 4.504,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 192.2848358154297,
      "learning_rate": 3e-05,
      "loss": 0.3724,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4105444550514221,
      "eval_runtime": 13.1508,
      "eval_samples_per_second": 285.155,
      "eval_steps_per_second": 4.486,
      "step": 2997
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}