trainer_state.json

{
  "best_metric": 0.7730666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0113/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": Infinity,
      "learning_rate": 0.0003,
      "loss": 2.2107,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6386666666666667,
      "eval_loss": 1.2155667543411255,
      "eval_runtime": 6.1642,
      "eval_samples_per_second": 608.355,
      "eval_steps_per_second": 9.571,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.918,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7077333333333333,
      "eval_loss": 0.9739989638328552,
      "eval_runtime": 5.6741,
      "eval_samples_per_second": 660.899,
      "eval_steps_per_second": 10.398,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.097511291503906,
      "learning_rate": 0.0003,
      "loss": 0.5962,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7248,
      "eval_loss": 0.9193477630615234,
      "eval_runtime": 6.1017,
      "eval_samples_per_second": 614.582,
      "eval_steps_per_second": 9.669,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 10.132493019104004,
      "learning_rate": 0.0003,
      "loss": 0.4213,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7613333333333333,
      "eval_loss": 0.8287147283554077,
      "eval_runtime": 5.6286,
      "eval_samples_per_second": 666.245,
      "eval_steps_per_second": 10.482,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.161423206329346,
      "learning_rate": 0.0003,
      "loss": 0.316,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7730666666666667,
      "eval_loss": 0.8246147036552429,
      "eval_runtime": 5.6332,
      "eval_samples_per_second": 665.694,
      "eval_steps_per_second": 10.474,
      "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}