trainer_state.json

{
  "best_metric": 0.8986666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0657/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.2736525535583496,
      "learning_rate": 8.657457896300791e-05,
      "loss": 1.6592,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7538666666666667,
      "eval_loss": 0.8866291046142578,
      "eval_runtime": 7.2246,
      "eval_samples_per_second": 519.062,
      "eval_steps_per_second": 8.167,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.685730736804761e-05,
      "loss": 0.5448,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8026666666666666,
      "eval_loss": 0.6555348038673401,
      "eval_runtime": 7.7161,
      "eval_samples_per_second": 485.999,
      "eval_steps_per_second": 7.646,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.9496603608131409,
      "learning_rate": 6.226977039554153e-05,
      "loss": 0.3214,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.45658496022224426,
      "eval_runtime": 7.8016,
      "eval_samples_per_second": 480.671,
      "eval_steps_per_second": 7.563,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.4970409870147705,
      "learning_rate": 4.505306743117272e-05,
      "loss": 0.187,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8749333333333333,
      "eval_loss": 0.4338541626930237,
      "eval_runtime": 7.7252,
      "eval_samples_per_second": 485.422,
      "eval_steps_per_second": 7.637,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.13811548054218292,
      "learning_rate": 2.7828285431465317e-05,
      "loss": 0.1065,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8816,
      "eval_loss": 0.4089241027832031,
      "eval_runtime": 7.2869,
      "eval_samples_per_second": 514.622,
      "eval_steps_per_second": 8.097,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.3568909764289856,
      "learning_rate": 1.3217741312837138e-05,
      "loss": 0.0518,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.39138007164001465,
      "eval_runtime": 7.3498,
      "eval_samples_per_second": 510.218,
      "eval_steps_per_second": 8.027,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.19655214250087738,
      "learning_rate": 3.4457579724370674e-06,
      "loss": 0.0286,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8936,
      "eval_loss": 0.4012141823768616,
      "eval_runtime": 7.4317,
      "eval_samples_per_second": 504.593,
      "eval_steps_per_second": 7.939,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.3163028657436371,
      "learning_rate": 3.12905914479078e-11,
      "loss": 0.0184,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8986666666666666,
      "eval_loss": 0.3862427771091461,
      "eval_runtime": 7.1478,
      "eval_samples_per_second": 524.638,
      "eval_steps_per_second": 8.254,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}