trainer_state.json

{
  "best_metric": 0.9389333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0035/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.0718417167663574,
      "learning_rate": 0.0002713525491562421,
      "loss": 0.6539,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8517333333333333,
      "eval_loss": 0.5032714605331421,
      "eval_runtime": 13.3012,
      "eval_samples_per_second": 281.928,
      "eval_steps_per_second": 4.436,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00019662164055386566,
      "loss": 0.2596,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.896,
      "eval_loss": 0.36434975266456604,
      "eval_runtime": 13.283,
      "eval_samples_per_second": 282.316,
      "eval_steps_per_second": 4.442,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.04782705008983612,
      "learning_rate": 0.00010391670726449014,
      "loss": 0.1153,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.924,
      "eval_loss": 0.2886028587818146,
      "eval_runtime": 13.2445,
      "eval_samples_per_second": 283.136,
      "eval_steps_per_second": 4.455,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0082821324467659,
      "learning_rate": 2.8814025486568238e-05,
      "loss": 0.0283,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9346666666666666,
      "eval_loss": 0.23317421972751617,
      "eval_runtime": 13.2876,
      "eval_samples_per_second": 282.218,
      "eval_steps_per_second": 4.44,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.05621960759162903,
      "learning_rate": 2.670129987614178e-10,
      "loss": 0.0086,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.23363636434078217,
      "eval_runtime": 13.4744,
      "eval_samples_per_second": 278.305,
      "eval_steps_per_second": 4.379,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}