trainer_state.json

{
  "best_metric": 0.9173333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0927/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": 10.243700981140137,
      "learning_rate": 8.753753753753754e-05,
      "loss": 0.8156,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8322666666666667,
      "eval_loss": 0.5370187759399414,
      "eval_runtime": 10.7202,
      "eval_samples_per_second": 349.806,
      "eval_steps_per_second": 5.504,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6557225584983826,
      "learning_rate": 7.507507507507507e-05,
      "loss": 0.2884,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.4544456899166107,
      "eval_runtime": 10.35,
      "eval_samples_per_second": 362.318,
      "eval_steps_per_second": 5.7,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.566348075866699,
      "learning_rate": 6.257507507507508e-05,
      "loss": 0.1462,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.4877142310142517,
      "eval_runtime": 10.4834,
      "eval_samples_per_second": 357.707,
      "eval_steps_per_second": 5.628,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.75417423248291,
      "learning_rate": 5.0075075075075074e-05,
      "loss": 0.0859,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8888,
      "eval_loss": 0.4447403848171234,
      "eval_runtime": 10.505,
      "eval_samples_per_second": 356.974,
      "eval_steps_per_second": 5.616,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.24770772457122803,
      "learning_rate": 3.7575075075075075e-05,
      "loss": 0.0438,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8912,
      "eval_loss": 0.479271799325943,
      "eval_runtime": 10.3038,
      "eval_samples_per_second": 363.944,
      "eval_steps_per_second": 5.726,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0001786894426913932,
      "learning_rate": 2.5075075075075076e-05,
      "loss": 0.0166,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.43210268020629883,
      "eval_runtime": 10.4688,
      "eval_samples_per_second": 358.207,
      "eval_steps_per_second": 5.636,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0004563328402582556,
      "learning_rate": 1.2575075075075077e-05,
      "loss": 0.0037,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9168,
      "eval_loss": 0.3666068911552429,
      "eval_runtime": 10.2807,
      "eval_samples_per_second": 364.759,
      "eval_steps_per_second": 5.739,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0008575678220950067,
      "learning_rate": 7.507507507507508e-08,
      "loss": 0.0009,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.3755517601966858,
      "eval_runtime": 10.6072,
      "eval_samples_per_second": 353.532,
      "eval_steps_per_second": 5.562,
      "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
}