trainer_state.json

{
  "best_metric": 0.8746666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0587/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": 2.401801801801802e-05,
      "loss": 2.3788,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7418666666666667,
      "eval_loss": 1.2080060243606567,
      "eval_runtime": 5.8368,
      "eval_samples_per_second": 642.473,
      "eval_steps_per_second": 10.108,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.368967056274414,
      "learning_rate": 1.801801801801802e-05,
      "loss": 0.853,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8218666666666666,
      "eval_loss": 0.7044945955276489,
      "eval_runtime": 5.9105,
      "eval_samples_per_second": 634.465,
      "eval_steps_per_second": 9.982,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 16.133573532104492,
      "learning_rate": 1.2018018018018019e-05,
      "loss": 0.5187,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.5509940385818481,
      "eval_runtime": 5.887,
      "eval_samples_per_second": 636.994,
      "eval_steps_per_second": 10.022,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.399643898010254,
      "learning_rate": 6.018018018018018e-06,
      "loss": 0.3705,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.49005332589149475,
      "eval_runtime": 6.4328,
      "eval_samples_per_second": 582.947,
      "eval_steps_per_second": 9.172,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 31.803041458129883,
      "learning_rate": 1.8018018018018017e-08,
      "loss": 0.2856,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.45511433482170105,
      "eval_runtime": 5.995,
      "eval_samples_per_second": 625.517,
      "eval_steps_per_second": 9.841,
      "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
}