trainer_state.json

{
  "best_metric": 0.9458666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0084/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": 0.39594656229019165,
      "learning_rate": 9.621650897886543e-05,
      "loss": 0.6536,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9184,
      "eval_loss": 0.2690630555152893,
      "eval_runtime": 11.4441,
      "eval_samples_per_second": 327.68,
      "eval_steps_per_second": 5.155,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.034524980932474136,
      "learning_rate": 8.539700818671957e-05,
      "loss": 0.1537,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9293333333333333,
      "eval_loss": 0.2537638247013092,
      "eval_runtime": 11.1129,
      "eval_samples_per_second": 337.444,
      "eval_steps_per_second": 5.309,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.24914999306201935,
      "learning_rate": 6.918863377282392e-05,
      "loss": 0.0614,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9368,
      "eval_loss": 0.25443124771118164,
      "eval_runtime": 13.4522,
      "eval_samples_per_second": 278.766,
      "eval_steps_per_second": 4.386,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.15276160836219788,
      "learning_rate": 5.005896381241413e-05,
      "loss": 0.026,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9352,
      "eval_loss": 0.26045969128608704,
      "eval_runtime": 13.2744,
      "eval_samples_per_second": 282.499,
      "eval_steps_per_second": 4.445,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.007646479178220034,
      "learning_rate": 3.0920317146072576e-05,
      "loss": 0.0072,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9448,
      "eval_loss": 0.2405226230621338,
      "eval_runtime": 10.6496,
      "eval_samples_per_second": 352.126,
      "eval_steps_per_second": 5.54,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.010624893009662628,
      "learning_rate": 1.4686379236485708e-05,
      "loss": 0.0023,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9445333333333333,
      "eval_loss": 0.24376659095287323,
      "eval_runtime": 10.614,
      "eval_samples_per_second": 353.308,
      "eval_steps_per_second": 5.559,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.02559749223291874,
      "learning_rate": 3.82861996937452e-06,
      "loss": 0.0011,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9437333333333333,
      "eval_loss": 0.24647106230258942,
      "eval_runtime": 10.7083,
      "eval_samples_per_second": 350.195,
      "eval_steps_per_second": 5.51,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.006798462010920048,
      "learning_rate": 3.476732383100867e-11,
      "loss": 0.001,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9458666666666666,
      "eval_loss": 0.24225540459156036,
      "eval_runtime": 10.4106,
      "eval_samples_per_second": 360.208,
      "eval_steps_per_second": 5.667,
      "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
}