trainer_state.json

{
  "best_metric": 0.8853333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0589/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": 2.651308536529541,
      "learning_rate": 0.0003,
      "loss": 0.6056,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.4713326692581177,
      "eval_runtime": 11.2851,
      "eval_samples_per_second": 332.296,
      "eval_steps_per_second": 5.228,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.3178,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.4441535472869873,
      "eval_runtime": 11.5819,
      "eval_samples_per_second": 323.781,
      "eval_steps_per_second": 5.094,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.13174070417881012,
      "learning_rate": 0.0003,
      "loss": 0.235,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8688,
      "eval_loss": 0.4864368736743927,
      "eval_runtime": 11.7318,
      "eval_samples_per_second": 319.645,
      "eval_steps_per_second": 5.029,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1318257451057434,
      "learning_rate": 0.0003,
      "loss": 0.191,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8666666666666667,
      "eval_loss": 0.4919469356536865,
      "eval_runtime": 11.4264,
      "eval_samples_per_second": 328.188,
      "eval_steps_per_second": 5.163,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.0423105955123901,
      "learning_rate": 0.0003,
      "loss": 0.1611,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.5422087907791138,
      "eval_runtime": 10.9072,
      "eval_samples_per_second": 343.811,
      "eval_steps_per_second": 5.409,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.021564319729804993,
      "learning_rate": 0.0003,
      "loss": 0.1426,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8853333333333333,
      "eval_loss": 0.4378768503665924,
      "eval_runtime": 10.9582,
      "eval_samples_per_second": 342.211,
      "eval_steps_per_second": 5.384,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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
}