trainer_state.json

{
  "best_metric": 0.8362666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0833/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": 7e-05,
      "loss": 1.7074,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7517333333333334,
      "eval_loss": 0.854519248008728,
      "eval_runtime": 4.4276,
      "eval_samples_per_second": 846.965,
      "eval_steps_per_second": 13.326,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.6115,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.808,
      "eval_loss": 0.6311678886413574,
      "eval_runtime": 4.2771,
      "eval_samples_per_second": 876.756,
      "eval_steps_per_second": 13.794,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.105736494064331,
      "learning_rate": 7e-05,
      "loss": 0.3804,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8165333333333333,
      "eval_loss": 0.5895447731018066,
      "eval_runtime": 4.2985,
      "eval_samples_per_second": 872.406,
      "eval_steps_per_second": 13.726,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.579101085662842,
      "learning_rate": 7e-05,
      "loss": 0.2564,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8165333333333333,
      "eval_loss": 0.5888329148292542,
      "eval_runtime": 4.2962,
      "eval_samples_per_second": 872.863,
      "eval_steps_per_second": 13.733,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.21133244037628174,
      "learning_rate": 7e-05,
      "loss": 0.1838,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8362666666666667,
      "eval_loss": 0.585121214389801,
      "eval_runtime": 4.5206,
      "eval_samples_per_second": 829.53,
      "eval_steps_per_second": 13.051,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}