trainer_state.json

{
  "best_metric": 0.924,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0189/checkpoints/checkpoint-999",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 999,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.39529821276664734,
      "learning_rate": 0.000375,
      "loss": 0.767,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8346666666666667,
      "eval_loss": 0.5728692412376404,
      "eval_runtime": 7.1124,
      "eval_samples_per_second": 527.25,
      "eval_steps_per_second": 8.295,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.043489716947078705,
      "learning_rate": 0.00012500000000000006,
      "loss": 0.2686,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9098666666666667,
      "eval_loss": 0.3065382242202759,
      "eval_runtime": 6.8617,
      "eval_samples_per_second": 546.516,
      "eval_steps_per_second": 8.599,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.19049251079559326,
      "learning_rate": 0.0,
      "loss": 0.0493,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.924,
      "eval_loss": 0.26637721061706543,
      "eval_runtime": 7.623,
      "eval_samples_per_second": 491.932,
      "eval_steps_per_second": 7.74,
      "step": 999
    }
  ],
  "logging_steps": 500,
  "max_steps": 999,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 4.94223961867776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}