trainer_state.json

{
  "best_metric": 0.4106666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0554/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "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": 0.0003,
      "loss": 3.2223,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2584,
      "eval_loss": 2.742262601852417,
      "eval_runtime": 10.6909,
      "eval_samples_per_second": 350.764,
      "eval_steps_per_second": 5.519,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.713911056518555,
      "learning_rate": 0.0003,
      "loss": 2.642,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.29173333333333334,
      "eval_loss": 2.5519378185272217,
      "eval_runtime": 11.0522,
      "eval_samples_per_second": 339.298,
      "eval_steps_per_second": 5.338,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 22.97842788696289,
      "learning_rate": 0.0003,
      "loss": 2.4233,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.33226666666666665,
      "eval_loss": 2.4291534423828125,
      "eval_runtime": 10.552,
      "eval_samples_per_second": 355.384,
      "eval_steps_per_second": 5.591,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 17.485658645629883,
      "learning_rate": 0.0003,
      "loss": 2.268,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.35146666666666665,
      "eval_loss": 2.3787593841552734,
      "eval_runtime": 10.4635,
      "eval_samples_per_second": 358.389,
      "eval_steps_per_second": 5.639,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 21.314083099365234,
      "learning_rate": 0.0003,
      "loss": 2.1337,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3752,
      "eval_loss": 2.2225289344787598,
      "eval_runtime": 10.6095,
      "eval_samples_per_second": 353.457,
      "eval_steps_per_second": 5.561,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 33.63254928588867,
      "learning_rate": 0.0003,
      "loss": 2.0188,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.4106666666666667,
      "eval_loss": 2.081040382385254,
      "eval_runtime": 10.6871,
      "eval_samples_per_second": 350.889,
      "eval_steps_per_second": 5.521,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}