trainer_state.json

{
  "best_metric": 0.9058666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0146/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8102321028709412,
      "learning_rate": 9.619397662556435e-05,
      "loss": 1.5527,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.808,
      "eval_loss": 0.6766788959503174,
      "eval_runtime": 6.3181,
      "eval_samples_per_second": 593.536,
      "eval_steps_per_second": 9.338,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.404699444770813,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.4624,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.4961727261543274,
      "eval_runtime": 6.1222,
      "eval_samples_per_second": 612.521,
      "eval_steps_per_second": 9.637,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.18833401799201965,
      "learning_rate": 6.918863377282392e-05,
      "loss": 0.2221,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.4809333384037018,
      "eval_runtime": 6.3114,
      "eval_samples_per_second": 594.16,
      "eval_steps_per_second": 9.348,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0659579038619995,
      "learning_rate": 5.005896381241413e-05,
      "loss": 0.107,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.4354960322380066,
      "eval_runtime": 6.3723,
      "eval_samples_per_second": 588.48,
      "eval_steps_per_second": 9.259,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.03774294629693031,
      "learning_rate": 3.0920317146072576e-05,
      "loss": 0.0388,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8912,
      "eval_loss": 0.40038129687309265,
      "eval_runtime": 6.247,
      "eval_samples_per_second": 600.293,
      "eval_steps_per_second": 9.445,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.047072041779756546,
      "learning_rate": 1.4686379236485708e-05,
      "loss": 0.0113,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.904,
      "eval_loss": 0.38641783595085144,
      "eval_runtime": 6.1837,
      "eval_samples_per_second": 606.429,
      "eval_steps_per_second": 9.541,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.2006550282239914,
      "learning_rate": 3.82861996937452e-06,
      "loss": 0.0042,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9058666666666667,
      "eval_loss": 0.37817421555519104,
      "eval_runtime": 6.1634,
      "eval_samples_per_second": 608.435,
      "eval_steps_per_second": 9.573,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}