trainer_state.json

{
  "best_metric": 0.9064,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0673/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 25.753623962402344,
      "learning_rate": 0.0004522542485937369,
      "loss": 1.2742,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8394666666666667,
      "eval_loss": 0.5269849896430969,
      "eval_runtime": 7.5711,
      "eval_samples_per_second": 495.302,
      "eval_steps_per_second": 7.793,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 35.782958984375,
      "learning_rate": 0.00032725424859373687,
      "loss": 0.3744,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.3974432349205017,
      "eval_runtime": 6.9976,
      "eval_samples_per_second": 535.898,
      "eval_steps_per_second": 8.431,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 30.030733108520508,
      "learning_rate": 0.00017274575140626317,
      "loss": 0.1655,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8917333333333334,
      "eval_loss": 0.3759390115737915,
      "eval_runtime": 7.2599,
      "eval_samples_per_second": 516.538,
      "eval_steps_per_second": 8.127,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 56.12133026123047,
      "learning_rate": 4.7745751406263163e-05,
      "loss": 0.0662,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9050666666666667,
      "eval_loss": 0.3498336970806122,
      "eval_runtime": 7.1905,
      "eval_samples_per_second": 521.52,
      "eval_steps_per_second": 8.205,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 46.435001373291016,
      "learning_rate": 0.0,
      "loss": 0.0376,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9064,
      "eval_loss": 0.3612825274467468,
      "eval_runtime": 6.9951,
      "eval_samples_per_second": 536.089,
      "eval_steps_per_second": 8.434,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}